Datasets:

eswardivi

/

books

like 0

easing diversity of personal devices get disconnected, broken, lost and generally get in the way of personal productivity. Getting rid of them has been the major motivation behind the efforts of the numerous working groups and other organisations involved since the late 1990s in developing standards for personal area networks. Within the IEEE, the 802.15 Working Group was established in March 1999 with the objective of providing standards to support the interoperability of low complexity, low power devices that can be worn, carried or located in a personal operating space (POS), defined as extending 10 metres in all directions around a stationary or moving person. An overview of the wireless PAN standards developed by the 802.15 Working Group is shown in Table 10-1. The kind of devices that were envisaged as participating in the PAN were the already ubiquitous mobile phone and then increasingly common personal stereo, pager and PDA. Considering that someone carrying a watch, mobile phone, pager, personal stereo and PDA would be equipped with two input keypads, four speakers, two microphones and potentially five LCD displays, the potential for simplification through interoperability is self-evident. A variety of different PAN standards have been developed since the late 1990s, most notably Bluetooth and IrDA, and more recently ZigBee and Wireless USB have come on the scene. Each of these technologies has its Chapter Ten Table 10-1: Overview of IEEE 802.15 WPAN Standards and Task Groups Standard Description Application 802.15.1 Original 2.4 GHz FHSS specification. Published in 2002. Bluetooth 802.15.2 Recommended practices to facilitate coexistence of 802.15 wireless PANs and 802.11 wireless LANs. Published 2003. 802.15.3a High rate WPAN. UWB PHY with DS-UWB VS OFDM under discussion. Draft published in 2003. Overtaken by MBOA and Wireless USB. Working Group disbanded in January 2006. 802.15.3b MAC amendment Task Group, improving implementation and interoperability of the 802.15 MAC. 802.15.3c Millimetre-wave altern

ative PHY. 57-64 GHz unlicensed band. 1 Gbps data rate and optionally to 2 Gbps. Formed in March 2005. 802.15.4 Low rate WPAN. DSSS 2.4 GHz, 915 and 868 ZigBee MHz. Published in 2003. 802.15.4a Task Group chartered to develop an alternative PHY layer. Two optional PHY specifications under consideration - a UWB impulse radio and a Chirp Spread Spectrum operating in the 2.4 unit space GHz ISM band. 802.15.4b Task Group chartered to address enhancements and clarifications to the 802.15.4 standard. 802.15.5 Developing MAC and PHY mechanisms required to enable mesh networking in wireless PANs. own particular strengths and weaknesses in the way it addresses the challenge of delivering easy to use. In this chapter the characteristics of each of these technologies will be described, from both a technical and a practical standpoint. Bluetooth (IEEE 802.15.1) Origins and Main Characteristics Research on the use of radio to link mobile phones and accessories was started by Ericsson Mobile Communications in 1994, but it was not until Wireless PAN Standards the Bluetooth Special Interest Group (SIG) was launched four years later, by Ericsson, IBM, Intel, Nokia and Toshiba, that the concept started to broaden beyond mobile phones to include connections between PCs and other devices. After the IEEE 802.15 Working Group was formed in 1999 with the task of developing standards for wireless PANs, the Bluetooth SIG was the only respondent to WG15's Call for Responses, and Bluetooth and IEEE 802.15.1 soon became synonymous. Since 2000, when Bluetooth-enabled wireless headsets started to emerge, cost and power usage have reduced significantly, and Bluetooth has become a common add-on feature for many mobile phones and PDAs. Bluetooth 1.1 is a PAN standard that operates in the 2.4 GHz ISM band at a PHY layer data rate of 1 Mbps, for an effective data rate of 721/56 kbps for asymmetric or 432 kbps for full duplex communication. Bluetooth 2.0 was ratified in November 2004, and enhanced data rate (EDR) was introduced which increased the

PHY data rate from 1 to 2 or 3 Mbps. A Bluetooth PAN can support up to eight devices in a piconet, with one device acting as master and up to seven as active slave devices. Piconets can be linked to form the so-called scatternets through the sharing of common devices, since a device can be both a master in one piconet and a slave in another. Each master device manages a piconet with a capacity of 720 kbps and a scatternet can have a much higher distributed capacity, under the control of multiple master devices. Bluetooth supports a wide variety of different types of devices and usage models, from mobile phone headsets to PDA synchronisation. Typically different usage models will call on different parts of the Bluetooth protocol stack. A profile is a vertical slice through the Bluetooth protocol stack (Figure 10-1), and represents the required protocols for a particular usage model. Profiles provide the basis for device interoperability and any given Bluetooth device may support a number of different usage models and therefore different profiles. Examples of the most important profiles are given in Table 10-2. Bluetooth achieves low component cost and extended battery life by settling for limited transmission range and modest data rates. Nevertheless, Chapter Ten Applications Protocols RFCOMM L2CAP Figure 10-1: Bluetooth Application Profiles Table 10-2: The Main Bluetooth Profiles Profile Description Personal area networking Enables general Internet protocol (IP) networking (including security) over an ad-hoc piconet. Synchronisation profile Enables the exchange of personal information such as calendar and address book data between devices. Basic printing profile Enables simple printing from a device to a printer. Specific printer drivers are not required in the sending device as the Bluetooth-enabled printer has the capability to decode the data sent to it to produce the required format. File Transfer Profile Enables a device to perform file management operations on another device's file system, including transfe

rring, creating or deleting files or folders. Headset profile Enables audio data transfer between a device such as a mobile phone or a PDA and a wireless headset. Dial-up networking profile Enables a dial-up networking link between a PDA or other device and a remote network. LAN access profile Enables a device to gain access to network resources such as storage or printers by using point-to-point protocol (PPP) to connect to another device that is already in a LAN. Wireless PAN Standards its effectiveness in common PAN tasks, such as telephony and short-range networking has resulted in it establishing a strong position among the available PAN options. Protocol Stack The Bluetooth protocol stack is illustrated in Figure 10-2. Above the Bluetooth radio (PHY layer), the Baseband, link manager protocol (LMP) and logical link control and adaptation (L2CAP) protocols correspond to the Data Link (LLC + MAC) layer of the OSI model. The following sections describe the PHY and Data Link layers, and outline the higher level protocols up to RFCOMM and service discovery protocol (SDP). The Bluetooth stack also included protocols adopted from other sources, such as OBEX an object exchange protocol adopted from IrDA (see the Section "IrDA Optional Protocol Stack, p. 284") and WAP (Wireless application protocol) - developed by the WAP Forum to support the delivery of Internet content over wireless links, primarily to mobile phones. Including WAP in the Bluetooth stack enables the reuse by Bluetooth profiles of software from the wireless application environment developed for WAP phones. Applications TCP/IP Bluetooth Host Protocol Stack RFCOMM (Software) Logical Link Control and Adaptation Protocol (L2CAP) Host Controller Interface (HCI) Host Controller Interface (HCI) firmware Bluetooth Host Controller Link Manager Protocol (LMP) Audio (Firmware and Hardware) Baseband / Link Controller (LC) Bluetooth radio (PHY) Figure 10-2: Bluetooth Protocol Stack Chapter Ten The Bluetooth Radio At the physical layer, Bluetooth uses the IEEE 80

2.15.1 radio, which specifies a frequency hopping spread spectrum system with 1600 hops per second between the 79 channels in the 2.40-2.48 GHz ISM band, and a hopping pattern controlled by the 48-bit MAC address of the master device. In some countries, the hopping pattern is reduced to cover just 23 channels in order to comply with specific local regulations. Gaussian frequency shift keying (GFSK) is used for the standard 1 Mbps PHY layer data rate (Bluetooth 1.2), while enhanced data rate (Bluetooth 2.0) uses t/4-DQPSK at 2 Mbps and 8-DPSK at 3 Mbps. The 2 Mbps rate is mandatory for Bluetooth 2.0 devices, while the 3 Mbps rate, relying on 8-DPSK which has a lower energy per transmitted bit and therefore a lower SNR (recall Eq. 4.1), is optional and only used over sufficiently robust links. To preserve backward compatibility with Bluetooth 1.2, GFSK is still used by Bluetooth 2.0 devices to transmit packet header information. Three classes of RF transmitted power are defined from 0 to 20 dBm (1-100 mW), as shown in Table 10-3. Most Bluetooth devices have class 3 radios, although class 1 adapters are also available, providing a PAN range comparable with an IEEE 802.11b/g wireless Table 10-3: Bluetooth RF Transmitter Power Classes Class Max RF power Range (ft) 100 mW (20 dBm) Up to 300 2.5 mW (4 dBm) Up to 30 1.0 mW (0 dBm) 0.3-3 Transmit power control is mandatory for class 1 radios (optional for class 2 and 3), and requires transmitting devices to dynamically adjust power in order to reduce interference. This also helps to reduce power consumption and extend battery life for portable devices. To implement power control a receiver signal strength indicator (RSSI) is used to determine whether a received signal is within a defined "golden receive power range", typically between 6 and 20 dBm above the receiver sensitivity level. If the received power is outside this power range, the receiver sends a link Wireless PAN Standards manager protocol (LMP) instruction to the transmitter to adjust its transmit power. Baseba

nd Layer The Bluetooth Baseband, sitting above the PHY layer in the Bluetooth protocol stack (Figure 10-2), manages the physical channels and links, including device discovery, link connection and management and power control. Time division multiplexing is used to divide access to the channel across devices in the piconet. Time is divided into slots of 650 us which are numbered according to the master device clock and allocated to links and devices by the master device. Two types of links are supported, synchronous connection-oriented (SCO) and asynchronous connection-less (ACL). SCO links mainly carry voice transmission data and are symmetric links between the master and a single slave. To maintain the link, the master reserves transmit/receive time slots at regular intervals and, since the link is synchronous, SCO packets are not retransmitted in the event of a packet error. ACL links connect the master to all the slave devices in the piconet (point to multi-point). The master device can establish an ACL link to any slave using time slots that are not reserved for any active SCO links. Only one ACL link can exist at a time, but the link can be to a slave that already has an SCO link to the master. For most ACL packets, packet retransmission is applied in the event of a packet error. The Bluetooth Baseband defines 13 packet types, including 4 specifically for high quality voice and voice + data transmission. Each packet consists of a 68-72 bit access code, a 54 bit header and a payload of up to 2745 bits. The access code is used during device discovery and to gain access to a specific piconet, and the header carries the slave address as well as information for acknowledgement, numbering and error checking of packets. The Baseband controls the process of device discovery through an inquiry procedure, which enables a device to discover other devices in range and determine their addresses and clock offsets, and a paging procedure which sets up the connection and synchronises the slave device clock to the master. On

ce a connection is established, a device can be in one of the Chapter Ten Table 10-4: Bluetooth Connection States State Description Active Devices in an active state participate in the channel. The active master schedules transmissions including regular transmissions to keep slaves synchronised. Active slaves listen for packets during ACL time slots. An active slave may sleep until the next ACL transmission if it is not addressed. Sniff Devices in the sniff state conserve power by listening for transmissions at a reduced rate. The inactive interval is programmable and will depend on the specific device type and application. A data transfer can be put on hold as a power saving measure either at the request of the slave device or under the direction of the master device. In the hold state only an internal timer will be running in the slave device. The data transfer will resume immediately after the slave device returns to active mode. Devices in the park state are still synchronised but do not participate in piconet traffic. Slaves give up their 3-bit active member device address when entering this state and take on an 8-bit parked member address. Parked devices will continue occasionally to listen to transmissions in order to re-synchronise and check on other broadcast messages. four states; active, sniff, hold, and park - in order of decreasing power consumption. Table 10-4 provides a brief description. Higher Layer Protocols Link Manager Protocol The link manager protocol (LMP) is used to set up and manage Baseband connections, including link configuration, authentication and power management functions. This is achieved by exchanging protocol data units (PDUs) between the Link Managers of two paired devices. PDUs include control of pairing, authentication, initiation of sniff, hold and park modes, power increase or decrease requests and selection of preferred packet coding and size to optimise data throughput. Host Controller Interface The host controller interface (HCI) provides a uniform command interface to t

he Link Manager and Baseband layers, allowing the protocol stack to Wireless PAN Standards be divided between two pieces of hardware - for example, a processor hosting higher layer software and a Bluetooth module. The host device performs upper layer functions and is able to interface with the second device performing LMP, Baseband and PHY layer functions. The two are connected through the Host Controller Transport layer, that can be either a UART, RS232 or USB interface. Logical Link Control and Adaptation (L2CAP) The logical link control and adaptation protocol (L2CAP) creates the logical connections between the upper layer protocols and the Baseband channels, assigning a channel identifier (CID) to each end-point of a channel. The process of establishing connections includes the exchange of information on the expected QoS between devices, and L2CAP monitors resource usage to ensure that QoS guarantees are met. L2CAP also manages the segmentation and re-assembly of data packets for higher layer protocols that use data packets larger than the Baseband's maximum transmission unit (MTU) of 341 bytes. RFCOMM The Bluetooth RFCOMM protocol is based on a subset of the ETSI TS 07.10 standard, and provides serial port emulation over the L2CAP protocol for cable replacement applications. RFCOMM assembles serial bit streams into bytes and data packets and provides reliable sequenced transport of the serial bit stream, using request to send/clear to send (RTS/CTS) and data terminal ready/data set ready (DTR/DSR) control signals. One adaptation of the ETSI standard implemented in RFCOMM is a credit-based flow control mechanism that limits frame transmission rate to ensure that the receiving device's input buffer does not overflow. If the credit counter for a connection reaches zero, RFCOMM will stop and wait until it receives more credit from the receiving device, indicating that the input buffer is able to receive data. The service discovery protocol (SDP) enables applications to discover what services are available on dev

ices in the piconet, and to determine the characteristics of available services. Chapter Ten Services are discovered using a request/response model, where an application sends out a protocol data unit requesting information on services available on a particular L2CAP connection, and awaits a response from the target device. Service discovery can be either by searching, requesting information on a specific desired service, or by browsing, requesting information on all available services. Bluetooth in Practice A Bluetooth piconet is established by the process of device discovery and pairing between master and discovered slave devices. This process can be repeated many times to create a PAN with up to 7 active slave devices, although 255 slave devices can remain connected to the piconet in the parked state. During the pairing process, the slave device receives a frequency hop synchronisation data packet, based on the master device's 48-bit MAC address, in order to follow the frequency hopping pattern. Once this low- level connection is made, the master device establishes a service discovery protocol (SDP) connection to determine which profile will be used to communicate with the slave device. LMP is then used to configure the link according to the specific service requirements. The two devices also exchange a passphrase which may be used to generate an encryption key to ensure secure communication, depending on security settings which are discussed further in Chapter 11. An example of a Bluetooth piconet, and the associated profiles used for the various device pairings is shown in Table 10-5. If the mobile phone in the above example is then paired with a Bluetooth enabled headset, it will become the master device in a second piconet, thus forming a scatternet. The mobile phone will then time-share between the two piconets. Time slots will be allocated to it by the master device in the first piconet and it will in turn allocate time slots to the headset and any other paired devices in the second piconet. Since the fr

equency hopping patterns of the two piconets are determined by different master device MAC addresses, they are not coordinated and there will be random data packet collisions when the same frequency is Wireless PAN Standards Table 10-5: Example of a Bluetooth PAN (Piconet) and Associated Profiles Device Master/Slave Profile Laptop computer Master Printer Slave Serial port profile Slave Synchronisation profile Laptop computer Slave File transfer profile Mobile phone Slave Dial-up networking profile chosen. However, this will occur very infrequently (statistically once in every 79 X 79 = 6241 data packets) and SO will not significantly affect data throughput. Bluetooth Usage Examples To give an idea of the way Bluetooth is used in practice it is useful to look at the steps involved in setting up a number of connections representing different profile usage. Synchronising a PDA over Bluetooth The steps in setting up this service, which uses the serial port and synchronisation profiles, are shown in Table 10-6. Table 10-6: Usage Model - PDA Synchronisation Description Pair the two devices (PDA and desktop or laptop), exchanging PIN/passkey. Use installed Bluetooth software on the host desktop/laptop to associate the incoming Bluetooth serial port with a specific COM port (e.g. COM3). Security of the link (encryption) will usually be specified in this step, and automatic synchronisation on connection may also be specified. Use synchronisation software on host desktop/laptop (e.g. Microsoft ActiveSync) to identify the serial port for synchronisation with the COM port specified in Step (2). Establish a connection with the host desktop/laptop from the PDA. Initiate synchronisation, unless automatic synchronisation on connection was selected at Step (2). Chapter Ten Internet connection from a PDA over Bluetooth The steps in setting up this service, which uses the LAN access profile, are shown in Table 10-7. Table 10-7: Usage Model - Internet Connection from a PDA Description Pair the two devices (PDA and desktop or laptop)

, exchanging PIN/passkey. Use installed Bluetooth software on the host desktop/laptop to configure network access in order to allow other devices to connect to the Internet/LAN through the host. Use the connection wizard or similar software on the PDA to establish a connection to the Internet through the host computer (rather than via a separate dial-up device such as a mobile phone). When the PDA attempts to connect to the host computer, authorise the connection from the host. If required, enter dial-up and/or log-on information to complete the connection. Dial-up Networking over a Bluetooth Enabled Mobile Phone The steps in setting up this service, which uses the dial-up networking profile, are shown in Table 10-8. Table 10-8: Usage Model - Dial-up Networking via a Bluetooth Enabled Mobile Phone Description Pair the two devices (the mobile phone and the computer requiring dial-up access), exchanging PIN/passkey. Use installed Bluetooth software on the computer to confirm via dial-up networking properties that a Bluetooth modem is installed for the mobile phone. If not this can be downloaded from the phone maker's Web site. On the computer, connect to dial-up networking (e.g. in Windows, double click on the dial-up networking icon in the Bluetooth devices folder). Enter username, password and ISP dial-up number when requested. Dial-up, user authentication and registration will follow. Wireless PAN Standards Application software such as dial-up networking or Internet connection may offer the option to save username and password information on the mobile device. This can be a security risk if the mobile device is lost and is not protected by a password. Bluetooth security, including a checklist of operational security measures, is discussed in the next chapter. Current and Future Developments Bluetooth's FHSS physical layer specification provides limited scope to compete with the higher data rates that will be available using wireless USB and other UWB radio based PAN standards. An increase in data rate to 3 Mbps

was achieved in Bluetooth 2.0, but much higher data rates are expected to be in demand by users with the ever growing volume of digital content being transferred between devices. In May 2005, the Bluetooth SIG announced that it is working with ultra wideband developers to extend Bluetooth capabilities to achieve the data rates that will be required by high speed applications such as delivering digital video to portable devices in a PAN. This development clearly aims to leverage the strong Bluetooth brand which has considerable market strength although, from the user's point of view, backward compatibility will be expensive to achieve as UWB and the current 2.4 GHz FHSS PHY are not interoperable. Wireless USB Origins and Main Characteristics Wireless USB is the result of the drive by the USB Implementers Forum to ensure that the highly successful wired USB interface evolves into the wireless future. The Wireless USB Promoter Group was formed in February 2004 to create a wireless extension to USB that would apply the original USB principles of ease-of-use, compatibility and low cost, to high speed wireless technology. The strong industry support and brand recognition of USB are assets that the group hopes will give wireless USB (WUSB) a head start in the wireless PAN sector. Retaining a strong link with wired USB is central to the design goals of wireless USB (Table 10-9). Wireless USB uses ultra wideband (UWB) radio technology to deliver a PHY layer data rate of 480 Mbps, with low power consumption and a range of up to 10 metres. This will enable wireless USB to comfortably Chapter Ten Table 10-9: Wireless USB Design Objectives Design goal Description Preserve USB Wireless USB is designed to use the same software interface software and device drivers as USB. infrastructure Preserve the smart As with USB devices, wireless USB keeps the device simple host simple device and leaves management of network complexity to the host. model Enable power Many devices that took power via a wired USB connection efficiency will b

ecome battery powered under wireless USB. Effective power management mechanisms will be required. Provide wired Device authentication and data encryption aim to provide the equivalent security same level of security as wired USB. Ease of use Plug and play is the user expectation for USB devices, and wireless USB is designed to continue that tradition. Preserve USB Wireless USB defines a specific device class (Wired Adapter) investment to allow wired USB devices or hosts to support wireless USB devices. stream video to multimedia consumer electronic devices, as well as offer high-speed connections to PC peripherals and other mobile devices. Revision 1.0 of the wireless USB specification was released in May 2005. Wireless USB devices share bandwidth through a host- scheduled TDMA based media access protocol. A hub and spokes logical topology is used, with each host supporting up to 127 devices and, like wired USB, system software is designed to accommodate devices connecting to or disconnecting from the host at any time. A "dual role" device is also defined to enable peer-to-peer connections. Protocol Stack The foundations of the wireless USB protocol stack are the PHY and MAC layer specifications developed by the MBOA-Special Interest Group (SIG). Since the merging of the MBOA-SIG and the WiMedia Alliance in March 2005, these specifications are now being finalised and promoted by the combined WiMedia-MBOA Alliance. Wireless PAN Standards Applications Wireless USB Other USB driver Wireless technologies Networking running over UWB radio Host Controller Interface Convergence Layer Medium Access Controller (UWB MAC) radio platform Ultrawideband radio (UWB PHY) Figure 10-3: Wireless USB Protocol Stack Wireless USB is one of the number of higher level technologies that will run over the MBOA PHY and MAC (Figure 10-3) layers, and the wireless USB specification defines the way in which a wireless USB communication channel is established using these lower layers. At the application level, wireless USB is functionally ident

ical to USB 2.0, except for some enhancements to the isochronous data communication model to allow for the relative unreliability of the wireless PHY layer compared to a wired USB connection. Wireless USB Radio The wireless USB PHY layer is the Multiband OFDM Alliance (MBOA) UWB radio, operating across the 3.1 to 10.6 GHz frequency bands. Support for data rates of 53.3, 106.7 and 200 Mbps is mandatory for wireless USB devices, with additional rates up to 480 Mbps being optional for devices and mandatory for hosts. Support for bands 1 through 3 (Channel 1 - see Figure 10-4) is mandatory for all wireless USB implementations, with optional support for other band groups. All time-frequency codes (TFC) for each band group supported must also be supported (see the Section "Multiband UWB, p. 122"). Media Access Control Layer The WiMedia MAC has been specifically developed to address the shortcomings of previous MACs, such as pre-802.11e Wi-Fi and Chapter Ten Channel 1 Channel 2 Channel 3 Channel 4 Channel 5 10296 528 MHz wide bands Figure 10-4: WUSB MBOA Bands Bluetooth, in providing guaranteed quality of service for real time video and audio streaming applications, as well as robustness to changing network topology. The key design features of the WiMedia MAC are summarised in Table 10-10. MAC layer timing is defined within the superframe structure shown in Figure 10-5. Each 65 ms superframe is divided into 256 media access slots (MAS) each of 256 us duration. The leading MASs in each superframe are used as a beacon period, during which devices exchange information with the host on their capabilities and resource requirements. Devices can reserve one or more medium access slots using distributed reservation protocol (DRP) messages during the beacon period. This enables applications to guarantee media access for isochronous data streams. Table 10-10: WiMedia MAC Key Design Features Design feature Description Distributed network At the MAC level, responsibility for medium control is control shared by all devices, reducing

vulnerability to single-point failure and eliminating the bandwidth penalty of maintaining central control. Prioritised access A TDMA system allows devices either to reserve mechanisms guaranteed medium access slots, or contend for access during a prioritised contention period. Network management The bandwidth overhead associated with the MAC protocol efficiency scales with the number of devices, assuring a low overhead for networks with few devices. Wireless PAN Standards 65 mS Super frame N-1 Super frame N Super frame N+1 Media access slots (MAS) Beacon period 256 us Figure 10-5: WiMedia MAC Superframe Structure Media access slots that are not reserved are available for use by any device, based on prioritised contention access (PCA). Here the prioritisation mechanism ensures that asynchronous but time-sensitive transmissions, such as those to and from a user interface device, will get priority over other non-time-sensitive devices. The contention access period is also used to increase the robustness of isochronous connections by providing an opportunity for the MAC layer to retransmit any failed packets from the DRP period. The Wireless USB Channel The wireless USB specification defines the way in which a wireless USB channel is established within the superframe structure and associated MAS reservation and control mechanisms. A host creates the wireless USB channel using DRP to reserve media access time slots that will be used for communication by all devices in the cluster. The host controls the channel using a sequence of control packets called micro-scheduled management commands (MMCs), which are transmitted during the reserved media access slots. These commands are used to dynamically schedule and control channel time for communication between the host and devices in the cluster. The MMC is a broadcast packet that contains a cluster ID to enable devices to identify control packets for their cluster. Each MMC specifies the breakdown into micro-scheduled channel time allocations (MS-CTAs) of reserved time un

til the next MMC. These allocations are used for data Chapter Ten communication within the cluster, with the direction and use of each MS-CTA being specified in the preceding MMC. As shown in Figure 10-6, although the MAC layer media access slots which support the channel will be discontinuous in time, the MMCs effectively pull these slots together into a contiguous channel for communication within the cluster. To ensure strict compliance with the MAC layer requirements, such as beaconing and distributed control, a wireless USB host is required to implement the full WiMedia MAC protocol. Other devices are only required to implement the wireless USB protocol that operates within the wireless USB channel. Wireless USB defines mechanisms to ensure that a host is aware of and respects the DRP reservations of any possible hidden neighbours of devices in its cluster. Three device classes are defined, as shown in Table 10-11, with different levels of "awareness" of the full MAC protocol. Directed beaconing devices have the capability to capture control protocol information transmitted by devices that are not part of the cluster and to transmit this back to the host via the wireless USB channel. This enables the host to periodically check for the presence of devices complying with the WiMedia MAC that are outside its range, and to ensure that MAC Super frame N-1 Super frame N Super frame N+1 DRP messages Reserved MASs Next MMC Next MMC Next MMC Transaction group 1 Transaction group 2 Transaction group n Figure 10-6: Wireless USB Channel within MAC Layer Superframes Wireless PAN Standards Table 10-11: Wireless USB Device Classes Device class Description Self-beaconing devices These devices comply fully with MAC level protocols and do all related beaconing. Directed beaconing devices These devices perform beaconing and other MAC functions under the direction of the wireless USB host device. Non-beaconing devices These devices have reduced transmit power and receiver sensitivity and can only operate in close proximity to th

e host. reservations made by the host and by non-cluster devices are mutually observable and respected. Directed beaconing devices must support three functions to enable these capabilities; A Count Packets function; by periodically counting packets during the beacon period, the host can determine whether a directed beaconing device has any hidden neighbours. A Capture Packets function; by capturing the beacon transmission of a hidden neighbour, the host can determine its DRP reservations, and adjust its own reservations if necessary. A Transmit Packet function; by providing the appropriate beacon data and instructing a directed beaconing device on when to transmit it, the host can inform the hidden neighbour of the presence and DRP reservations of nearby devices in the cluster. These control mechanisms also ensure that several wireless USB clusters can spatially overlap with minimum interference. The wireless USB protocol defines the packet formats and controls the various types of data transfers (bulk, isochronous and control) within the micro-scheduled wireless USB channel. The protocol also provides control of a range of measures designed to mitigate the effect or RF interference on transfer reliability, including control of transmit power, bit rate and size of the data payload within a transmitted packet, as well as bandwidth and RF channel switching. Chapter Ten Wireless USB in Practice One of the design goals of wireless USB is that it should preserve the "plug and play" ease of use associated with wired USB. The equivalent wireless concept of "turn on and use" will require wireless USB devices to automatically install drivers and security features when turned on for the first time, as well as identifying and associating with other devices with a minimum of user input for authentication. Besides the hub and spokes topology that allows a host to control up to 127 end-point devices, wireless USB allows a dual role device (DRD) to function as both a host and a device simultaneously. This allows simple peer-to-

peer connections between two dual role devices, with each device in its host role managing a separate wireless USB channel (called the default and reverse links) over a common MAC layer channel. DRDs can also be connected to one or more wireless USB channels as a device while at the same time providing a wireless USB channel for other devices as a host. An example of this combination scenario would be a wireless USB printer acting as a device to a laptop computer and as a host to a digital camera. Although wireless USB devices are still under development, several features of the specification, particularly of the MAC layer, point to some important characteristics that will have practical implications. The WiMedia PHY and MAC have features that allow the distance between devices to be determined by measurement of the two-way transfer time (TWTT) of a message between devices. Simple MAC to MAC transactions will allow devices to exchange measurement frames and pass a distance calculation up to the application layer. With several devices distributed in 3D, triangulation can then be used to establish the spatial location of each device. This can be put to a variety of uses, from the trivial (helping to find a lost PDA) to more significant location specific services. Current and Future Developments Although wireless USB is still in its infancy, it seems likely that, building on the foundation established by wired USB and the MBOA UWB radio platform (which is likely to be common to several other technologies including W1394), wireless USB will quickly become established as a Wireless PAN Standards standard for wireless interconnection of PC peripherals and other consumer electronic devices. The wireless USB architecture and protocol are scalable to higher data rates and, as the MBOA UWB radio platform evolves, data rates of 1 Gbps and beyond are likely to be achieved. ZigBee (IEEE 802.15.4) Origins and Main Characteristics ZigBee is a standards based technology addressing the needs of remote monitoring and control, and

sensory networks. The ZigBee Alliance was formed in November 2002 with the aim of exploiting the IEEE 802.15.4 radio, and the ZigBee 1.0 specification was finalised in March 2005. ZigBee delivers low data rate communications, up to 250 kbps, with ultra-low power consumption, and aims to provide a device control channel, rather than the high rate data flow channel targeted by technologies such as Wireless USB. Some of the specific steps that have been taken to achieve ultra-low power consumption are; reducing the amount of data that is transmitted including the frame overhead (addressing and other header information) reducing the transceiver's duty cycle including power management mechanisms for power-down and sleep modes targeting a limited operating range of around 30 metres. As a result, a ZigBee network will typically require only 1% of the power of an equivalent Bluetooth PAN, resulting in a battery life of months to years. Two device classes will be defined; full function devices which implement the full protocol stack and are capable of being co-ordinating nodes and connecting with any type of device in any topology, and reduced function devices which will implement a simplified protocol set and will be limited to acting as end-nodes in simple connection topologies (star or peer-to-peer). Every ZigBee network has a unique full function personal area network co-ordinator (similar to a Bluetooth Master device) which is responsible for network management tasks such as associating new devices and Chapter Ten beacon transmissions. In a star network all devices communicate with the PAN co-ordinator, while in a peer-to-peer network individual full function devices are able to communicate with each other. The very low target cost ($1-$5) will make ZigBee ideal for wireless monitoring and control applications such as residential and commercial building automation (smart home) and industrial process control. In the home ZigBee can enable the creation of a home area network (HAN), allowing the many devices currently c

ontrolled by a proliferation of non-interoperable remote controllers to be brought under the command of a single control unit. Protocol Stack The ZigBee 1.0 specification includes an upper layer protocol stack (Figure 10-7) building on the IEEE 802.15.4 PHY and MAC layer specifications that were finalised in May 2003. Logical network control, security and applications layer are optimised for time-critical applications, with very fast device wake-up and network association times, typically in the region of 15 and 30 ms respectively. ZigBee Application Framework Application Support Layer (APS) APS security Message Endpoint manager formatting multiplexing Network Application Layer (NWK) NWK security NWK message Routing Network manager broker manager manager 802.15.4 LLC 802.2 Type 1 LLC Logical Link Control Logical Link Control MAC (802.15.4) PHY (802.15.4) 868 / 915 MHz 2.4 GHz Figure 10-7: ZigBee Outline Protocol Stack Wireless PAN Standards Star links ZigBee ZigBee end devices coordinator ZigBee Mesh links end devices ZigBee router Figure 10-8: ZigBee Supported Topologies The network layer is responsible for the usual tasks of network start-up, associating, dissociating, address assignment to devices, security, frame routing, etc. Multiple network topologies are supported by the network layer, as shown in Figure 10-8. The mesh topology can extend a network to up to 64,000 nodes through the use of ZigBee routers, which achieve efficient routing using a request-response algorithm rather than the use of router tables. The general operating framework (GOF) is an integration layer linking the application and network layers, and maintains an overview of device descriptions and addresses, events, data formats and other information that is used by the applications to command and respond to networked devices. Finally, at the top of the stack and similar to Bluetooth, application profiles are defined to support specific usage modes. For example, a lighting profile would include sensors for light level and occupancy as well

Media Access and Link Control Layer ZigBee will use the IEEE 802. 15.4 MAC with 15.4a modifications, to support up to 64,000 nodes in a variety of simple connection topologies. In an extended network, device access to the physical channel is controlled using a combination of TDMA and CSMA/CA. A "superframe" structure (Figure 10-9) breaks the time period between beacon transmissions into 16 time slots. Beacons are transmitted by the PAN co-ordinator at predefined intervals of between 15 ms and 252 seconds, and are used to identify and synchronise devices in the network. These beacon messages are sufficiently infrequent not to suffer collisions, and are therefore not subject to CSMA/CA. The 16 time slots are divided into two access periods, a contention based access period when devices will use CSMA/CA to determine periods when they are able to transmit, and a contention free period when devices will use guaranteed time slots, assigned by the PAN co-ordinator (TDMA). The Wireless PAN Standards Super frame N-1 Super frame N Super frame N+1 Contention based access period 15 mS Inactive Beacon Optional guaranteed time slots Beacon Figure 10-9: ZigBee Superframe Structure combination of predefined beacon intervals and guaranteed transmission time slots allow sensing devices to conserve power during extended sleep periods, waking only to check in on beacons or to use a guaranteed transmission slot. CSMA/CA used during the contention access period is very similar to that used in IEEE 802.11 networks. Devices listen before transmitting and backoff a random number of time slots before attempting to transmit. The backoff period is doubled each time a collision is sensed. A transmitting device can set a header bit to request an acknowledgement (ACK) for a message and will retransmit a message a fixed number of times if the ACK is not received. A non-beacon mode is also defined for application in which the controller does not need to conserve power and where device transmission is SO infrequent that a collision avoidance stra

tegy is an unnecessary overhead. An example would be a security system, with a mains powered controller and (hopefully) very infrequent security alerts. Three device types are recognised in the IEEE 802.15.4 MAC specification; full function devices, the network co-ordinator that is a special type of full function device, and reduced function devices. These device types are described in Table 10-13. Devices use either full 64-bit or short 16-bit addresses, and transmitted frames can include both destination and source addresses. This is necessary for peer-to-peer connections and is also important in mesh topologies, to provide robustness to single point failures in the network. Chapter Ten Table 10-13: ZigBee Device Types Device type Description Full function device (FFD) FFDs carry all features specified by the IEEE 802.15.4 standard. They have additional memory and computing power to perform network routing functions and can be used as edge devices where the network interacts with external devices. Network co-ordinator The PAN co-ordinator is the most complex of the device types with the largest memory and computing power. It is a full function device that maintains overall control of the network. Reduced function device RFDs have limited functionality in order to reduce (RFD) complexity and cost. They can only communicate with FFDs and will generally be used as network edge devices. ZigBee in Practice ZigBee networks will cover a variety of short-range, low data rate applications from PC peripheral interfacing to industrial control, as shown in Table 10-14. A typical ZigBee home automation network might cover the first three of these application areas, bringing together control of lighting, security, home entertainment and a variety of PC peripherals into a single network. Table 10-14: ZigBee Application Areas Application area Application examples PC peripherals Mouse, keyboard and joystick interfaces Consumer electronics Home entertainment system (TV, VCR, DVD, audio system) remote control Residential and othe

r Security and access control, lighting, heating, ventilation building automation and air conditioning (HVAC), irrigation Healthcare Patient monitoring, fitness monitoring Industrial control Asset management, industrial process control, energy management. Wireless PAN Standards Although a ZigBee network may be competing for access to the 2.4 GHz ISM band with Wi-Fi and Bluetooth networks, as well as a range of other control and communication devices, ZigBee is likely to be very robust to potential interference since the duty cycle of a ZigBee device will generally be very low. The CSMA/CA mechanism, together with backoff and retry if no acknowledgement is received, means that if there is interference, ZigBee devices can simply wait for an opening and keep trying until packet reception is confirmed. The low duty cycle and low data volumes also mean that ZigBee devices are unlikely to cause significant interference to overlapping Wi-Fi or Bluetooth networks. Mesh Implementation Considerations A number of special considerations arise in order to ensure the successful implementation of a functional, robust and reliable ZigBee mesh network, for example in an extensive building or industrial automation application. Mesh functionality requires that every device is able to communicate with at least one, and preferably several other devices, providing one or more paths to the central controller or point of exit from the mesh. Clearly this is also the case for WLAN installations, but in a ZigBee mesh, where devices may be installed in or concealed by machinery or pipework, special attention to signal strength site surveying will be required. Weak or broken links in the mesh can result in the separation from the main mesh of one (orphan) or a group of devices (split mesh). The robustness of a mesh will also be adversely affected by any susceptibility to single link failures. During planning, the mesh topology should be carefully reviewed for single point vulnerability and after installation the signal strength of each link

should be checked to ensure that it is sufficient, particularly at aggregation points where data throughput is expected to be high. In extensive installations, with many tens or hundreds of installed devices, maintaining good records of device location and service history will also be important to ensure that the reliability of the network can be efficiently maintained. Current and Future Developments ZigBee is one of the number of competing technologies in the area of sensor networking and remote control. ZigBee's strengths are the IEEE Chapter Ten Table 10-15: Sensor Networking Technologies - Specific Strengths Technology/Alliance Technology strengths and key features Insteon Combination of 132 kHz power line modulation and wireless networking (915 MHz ISM band). Backward compatible with X-10 home automation systems. Dust Networks Full mesh networking - every device has message routing capability. Proprietary 25 channel FHSS radio in the 915 MHz ISM band. Z-Wave Full mesh networking. Proprietary 868/915 MHz ISM band radio (9.6 kbps, BFSK modulation) as well as 802.15.4 compliant products. standard on which it is based and the broad industry alliance which will assure interoperability over a wide product range. Competing sensor networking technologies, such as those offered by Dust Networks, Millennial Net and Insteon are proprietary, although Z-Wave is also backed by an industry alliance and aims to develop IEEE 802.15.4 based products. Each of these proprietary technologies has its own specific advantages, some of which are shown in Table 10-15. Future developments from the ZigBee Alliance could include a ZigBee 2.0 specification based on the enhanced low data rate specification currently under development by the IEEE 802.15 TG4a. This task group is working on an alternate physical layer specification for the 802.15.4 standard which aims to deliver a location capability with an accuracy of 1 metre or better, higher data throughput, ultra-low power and longer range, as well as lower cost. Origins and Main Chara

cteristics The Infrared Data Association (IrDA) began life in 1993 as a non-profit organisation with the aim of promoting the use of infrared communication links between PCs and other devices by developing and supporting standards to ensure hardware and software interoperability. The group Wireless PAN Standards published its first standard in June 1994, including the specification of the Serial Ir Link (SIR) which uses infrared to replace the serial interface cable. Since then IrDA has grown to be the most widespread wireless connection technology, with over 250 million IrDA compliant interfaces shipped in 2004. IrDA is a low-cost, low-power, serial data connection standard that supports a half-duplex, point-to-point connections with a range of at least one metre and a data rate of up to 115 kbps (SIR and standard power mode). IrDA operates at a wavelength of roughly 1 um compared to 12.5 cm for Bluetooth at 2.4 GHz. Unlike the omnidirectional coverage achievable with RF transmitters, IrDA's point-to-point connection model requires the Ir transceivers to be aligned within +30° in order for the receiver to be illuminated with the required minimum power density (Figure 10-10). This physical requirement makes IrDA well suited for some applications, like secure simple object exchange, but not SO well suited for others, such as ad-hoc networking or supporting audio or telephony headsets. IrDA has also been successful in developing generic protocols, such as the OBEX (object exchange) protocol, which allows devices to exchange objects such as business cards, files, pictures and calendar items. This was introduced by IrDA in 1997, and has been broadly adopted as a simple solution for object exchange over various transport options, including TCP/IP and Bluetooth. Half angle 15° OK; angular misalignment but within beamwidth enabled device Not OK; angular alignment but outside beam width Figure 10-10: IrDA Device Alignment Chapter Ten Protocol Stack The IrDA protocol stack supports data link initialisation and shutdown, c

onnection start-up and disconnection, device address discovery and conflict resolution, data rate negotiation and information exchange. On top of the PHY layer specification, IrDA has two mandatory protocols at the MAC level, as well as a number of optional layers that are available for specific usage models. The mandatory protocols are the link access (IrLAP) and link management (IrLMP) protocols (Figure 10-11). IrDA PHY Layer The IrDA infrared physical specification (IrPHY) covers aspects of the infrared beam such as wavelength, minimum and maximum power levels or irradiance in mW/sr (milliwatts per steradian) and beam angle, as well as the physical configuration of the optical components. Infrared wavelengths of 0.85-0.90 um are specified, since light emitting diodes and optical detectors for these wavelengths are readily available and at low-cost. Two power modes are specified - standard and low-power. Link distances of up to 0.2 m are possible in low-power mode, with a maximum power intensity of 28.2 mW/sr or up to 1m in standard mode with 500 mW/sr maximum power intensity. The IrPHY specification also defines the encoding and framing of data for various transmission speeds, as summarised in Table 10-16. SIR is an asynchronous format that uses the same data format as the standard UART (1 start bit, 8 data bits, 1 stop bit). An RZI modulation Network layer Link Management Protocol (IrLMP) Data Link layer Link Access Protocol (IrLAP) Physical layer Physical layer (IrPHY) OSI Model layers IrDA Mandatory protocols Figure 10-11: IrDA Mandatory Protocol Stack and the OSI Model Wireless PAN Standards Table 10-16: IrDA Data Rates and Modulation Methods Transmission type Data rate Modulation SIR (Serial Ir) 9.6-115.2 kbps FIR (Fast Ir) 0.576-1.152 Mbps 4 Mbps VFIR (Very Fast Ir) 16 Mbps method is used (Figure 10-12), with a short Ir pulse being transmitted for each zero data bit. The Ir pulse is shortened to nominally 3/16th of the bit duration in order to reduce the LED power consumption. FIR and VFIR are synchronou

s transmission formats using RZI, 4-PPM or HHH modulation methods. The Ir pulse durations are nominally 25% of the bit or symbol duration. All transmissions start at the lowest data speed of 9.6 kbps in order to ensure interoperability, and higher data rates are negotiated, depending on the capabilities of the communicating devices, as part of the process of establishing the link. Data Link Layer IrDA's two mandatory protocols, the link access (IrLAP) and link management (IrLMP) protocols, are Data Link (Layer 2) protocols in terms of the OSI model (Figure 10-11). Clock 16 X bit rate Data bits (NRZ) Return to zero inverted (RZI) Bit period 7 off 6 off 16 clock cycles per bit period Figure 10-12: IR Pulse Shortening (SIR RZI Modulation) Chapter Ten IrLAP establishes the device to device connection, controlling the discovery and addressing of devices within range and establishing the best common data transmission rate. On discovery, devices randomly choose and exchange 32-bit IrLAP addresses. The devices in an IrLAP connection have a master-slave relationship with the master responsible for sending command frames, initiating connections and transfers, organising and controlling the flow of data, including dealing with data link errors. The slave device sends response frames, responding to the commands and requests of the master device. Once the IrLAP connection is started, media access is controlled by time division (TDMA) with master and slave taking alternate 500 millisecond time slots. The distinction between master and slave devices is relevant at the Data Link level. However, at the application level, once a connection is made between two devices, an application on a slave device can initiate an operation on the master device just as easily as the other way round. IrLMP, the link management protocol, multiplexes services and applications on the connection established by IrLAP. IrLMP also handles address conflict resolution in the case of a new device being discovered which requests the same IrLAP address. IrDA

Optional Protocol Stack IrDA's optional protocol stack (Figure 10-13) offers applications a number of new services and emulations of legacy services, the most important of which are described below. LM-IAS The link management information access service (LM-IAS) provides a database to enable applications to discover devices and to access device-specific information, in essence a "yellow pages" of devices and the services they can provide. All services or applications available via an incoming connection must have an entry in the LM-IAS database, which can be queried to obtain information about these services, for example the current load experienced by a network resource or the attributes of a serial link emulation. Wireless PAN Standards Tiny-TP Tiny-TP is an intermediate protocol layer that provides a simple transport protocol to control flow on IrLMP connections. It also provides a segmentation and re-assembly service to prevent deadlock situations that can occur as a result of limited device buffer space. Tiny-TP controls flow by adding a single "credit" byte of overhead to each transmitted frame. A receiving device can use this credit when one of its applications need to transmit an LMP frame back to the other device. This simple system is similar to the flow control mechanism used in Bluetooth's RFCOMM protocol, described above, and ensures that communication is not interrupted by a device running out of buffer space. IrCOMM IrCOMM emulates legacy serial (or parallel) port connections for applications such as printing. When installed, IrCOMM creates a virtual port which appears to the host computer or applications as if it were a standard serial or parallel port connection. IrCOMM includes emulation of a number of legacy interfaces including RS232 and Centronics LPT. IrOBEX IrOBEX is an optional application layer protocol that is designed to enable applications to exchange a wide variety of arbitrary data objects such as files, electronic business cards and digital images. It defines the Information IrLAN I

rOBEX IrCOMM Access Service (LM-IAS) Tiny Transport Protocol (Tiny-TP) In Link Management Protocol (IrLMP) In Link Access Protocol (IrLAP) Physical layer (IrPHY) SIR 9.9-115.2 kbps FIR to 4 Mbps VFIR to 16 Mbps Figure 10-13: IrDA Optional Protocol Stack Chapter Ten conversion of any file into a universal object and also provides tools to enable the object to be understood and handled correctly on the receiving side of the link. IrOBEX serves a similar role to HTTP in the Internet protocol suite. IrLAN IrLAN allows devices to access Local Area Networks by emulating a low level Ethernet link, including TCP/IP. Using IrLAN, a computer can attach to a LAN via an access point device (an IrLAN adapter) or through a second computer already attached to the LAN. Two computers can also use IrLAN to communicate as though attached via a LAN, giving each computer access to the other computer's directories and other network resources. IrDA in Practice IrDA is the most widespread wireless networking technology in use today. It provides a simple and secure method for transferring files between personal computing and communication devices, and is firmly associated with such applications as PDA to laptop synchronisation, business card and mobile phone data exchange. Apart from IrDA ports in laptops and PDAs, over 200 million IrDA enabled mobile phones were shipped in 2004. With the increased popularity and pixel count of digital cameras in mobile phones, these IrDA links are also being put to use for direct photo printing and image file transfer. Current and Future Developments of IrDA The IrDA IrBurst and UFIR Special Interest Groups have been working since 2003 on the next generation IrDA specification, with IrBurst targeting 100 Mbps and Ultra Fast IR (UFIR) aiming for a data rate of 500 Mbps. These specifications will also deliver a new Ir protocol stack, since tests show that the existing IrCOMM and Tiny-TP protocols have maximum throughputs in the region of 3 Mbps. The market driver for these developments is seen as a user d

emand to transfer compressed video between devices, and the target is to transfer Wireless PAN Standards one hour of MPEG2 compressed video (100-200 MB) over a handheld link in no more than 10 seconds. One usage model anticipates that a customer will use a handheld device such as a mobile phone to pay for and download video content from vending machines on the street. Future developments will also see an extension of the effective range of IrDA beyond the current one metre limit. This will enable the "mobile phone as digital wallet" usage model to be extended out of doors, to applications such as motorway toll collection. Near Field Communications Origins and Main Characteristics Near field communication (NFC) is an ultra short range wireless communication technology that uses magnetic field induction to enable connectivity between devices when they are in physical contact or within a range of a few centimetres. NFC has emerged as a technology for interconnecting consumer electronic devices from the convergence of contactless identification (e.g. RFID) and networking technologies, and aims at simple peer-to-peer networking through automatic connection and configuration. The key difference between NFC and standard RF wireless communication is the way in which the RF signal is propagated between the transmitter and receiver, as described in the Section "RF Signal Propagation and Reception, p. 105". Standard RF communications, such as a Wi-Fi, is described as "far-field" since the communication range is large compared to the size of the antenna. Near field communication relies on direct magnetic or electrostatic coupling between components within the communicating devices rather than free space propagation of radio waves. Because of the very short range, NFC devices can communicate using extremely low electric or magnetic field strengths, well below regulatory noise emission thresholds, SO that there are no limitations on frequency band usage due to licensing restrictions. NFC technology is a joint development of Ph

ilips and Sony, and is based on the ECMA 340 standard. The technology is being promoted by the NFC Forum, whose sponsor members also include MasterCard, Motorola, Nokia and Visa International. Chapter Ten The ECMA 340 standard was adopted by the ECMA General Assembly in December 2004, and defines NFC communication modes using inductive coupled devices operating at a centre frequency of 13.56 MHz. The definition is also known as the near field communication interface and protocol (NFCIP-1). Similar to the more familiar IEEE standards, ECMA 340 specifies the modulation and data coding schemes, data rates and frame format for NFC device interfaces. A simple link layer protocol addresses link initialisation and collision avoidance, and a transport protocol, covers protocol activation, data exchange and deactivation. NFC PHY Layer ECMA 340 specifies a magnetic induction interface operating at 13.56 MHz and with data rates of 106, 212 and 424 kbps, which is compatible with Philips' MIFARE® and Sony's FeliCaTM contactless smart card interfaces. Rather than measuring transmitter power and receiver detection levels in dBm as is the case for far-field RF communication, the strength (H) of the magnetic field used in NFC is measured in amps/metre (A/m). ECMA 340 specifies the field values as shown in Table 10-17. The ECMA 340 standard defines two communication modes - active and passive. In the active mode, communication is started by an RF field generated by the initiating device (the Initiator) and the target device (the Target) also generates a modulated RF field to respond to the Initiator's command. Modulation and bit coding methods used in active mode are shown in Table 10-18. In the passive mode (Table 10-19), the Initiator starts the communication using an RF field but the Target responds by load modulation rather than by generating an RF field in response. Load modulation, described in the Table 10-17: ECMA 340 NFC Magnetic Field Strength Specification Field level Field strength Description Hthreshold 0.1875 A/m Min

imum field detection level 1.5 A/m rms Minimum un-modulated field strength 7.5 A/m rms Maximum un-modulated field strength Wireless PAN Standards Table 10-18: ECMA 340 Active Mode Modulation and Bit Coding Methods Bit rate Modulation Bit coding method method 106 kbps ASK (100% Pulse position coding (Modified Miller) - modulation) pulse transmitted at the centre of a bit period for each 1-bit, or at the start of a bit period for an opening 0-bit or a repeated 0-bit. 212/424 kbps ASK (8-30% Manchester coding - transition at the centre modulation) of each bit period; low to high for a 0-bit, high to low for a 1-bit. Reverse polarity (i.e. high to low for a 0-bit, low to high for a 1-bit) is also allowed. Section "Load Modulation, p. 127", entails modulating the load in the target device that the initiating RF field is applied to. This generates sidebands on the original carrier frequency (13.56 MHz) that are detected by the Initiator. Protocol Stack Since NFC is not attempting to provide the full range of network features captured in the OSI model, the protocol stack is very limited and consists of a single simple transport protocol, which defines activation, data exchange and deactivation on an NFC link. The vestiges of a Data Link layer are also evident in the form of media access control based on CSMA/CA. An Initiator checks for an existing Table 10-19: ECMA 340 Passive Mode Modulation and Bit Coding Methods Bit rate Modulation Subcarrier Bit coding method method frequency 106 kbps f/16 = 847.5 kHz Subcarrier modulated using modulation Manchester coding. Reverse polarity not allowed. 212/424 kbps Carrier modulated using modulation Manchester coding. Reverse polarity allowed. Chapter Ten RF field before commencing communication and similarly a Target device in active mode checks for an existing RF field before responding. A single initiating device can interact with multiple targets, each of which generates a random 40 bit ID at the start of the device selection process. The discovery of target device IDs involves

an elegant process to resolve collisions which will occur when several targets respond at the same time, particularly when targets are responding in passive mode (Figure 10-14). Collision detection at the bit level is made possible by the use of Manchester coding, since a collision is detected when a full bit period occurs without a transition being sensed. This can only occur when a 1-bit transmitted by one target collides with a 0-bit transmitted by another target. Bits received before the collision can be recovered and the targets are requested to re-send data starting with the unrecovered bit. A random delay used by responding targets ensures that this process does not get stuck in a repeating loop. The data link between devices is transaction based, with initiation and termination occurring around a single data transfer. Initiator and Target negotiate a communication speed, starting with the lowest (106 kbps), in a parameter selection step during transport protocol initiation. Data bits (Responder 1) Random retransmit delay Data bits (Responder 2) Random retransmit delay Combined response Manchester coded Detected bits (Initiator) First two bits Initiator transmits Next two bits Resend successfully Collision resend request successfully Collision request detected detected detected detected Figure 10-14: NFC Collision Detection with Multiple Responding Targets Wireless PAN Standards NFC in Practice Four basic NFC usage models are currently envisaged, as shown in Table 10-20. Apart from these usage models in which the NFC connection is used to transfer end-user data, NFC can also be used to securely initiate another connection between two NFC enabled devices. For example, NFC Table 10-20: NFC Usage Models Usage model Description Example Touch and go The user brings the device You see a poster advertising an storing a ticket or access event such as a concert you code close to the reader for want to attend. Bring your applications such as event or PDA or mobile phone near the transport ticketing and access poste

r to download event control, or for simple data information from a smart chip capture, such as picking up in the poster. an Internet URL for further information from a smart label on a poster or other advertising. Touch and Transactions such as mobile Event tickets could be confirm payment where the user is purchased online or from an required to enter a password electronic box office and or other confirmation to stored on your handheld authorise the interaction. device. Touch and Two NFC-enabled devices If you take pictures with your connect can be linked to enable peer- mobile phone's built in to-peer data transfers, such camera, you can later touch an as exchanging photos or NFC enabled computer or TV synchronising contact to display the images, or touch information. and transfer them to a friend's mobile phone. Touch and NFC enabled devices may offer By simply touching two explore a range of possible functions, devices together it will be including other high speed possible to transfer large files connectivity options. The simple between the devices, for NFC connection will allow example, using NFC to identify the user to explore a device's and configure a separate capabilities and access other high-speed wireless connection. available services or functionality. Chapter Ten enabled Bluetooth or Wi-Fi devices may use NFC to initiate and configure the longer range link. Security is assured by the close proximity requirement for NFC operation. Once the Bluetooth or Wi-Fi link is configured, the devices can be separated for longer range communication. Current and Future Developments of NFC The first examples of NFC in use have been trials followed by commercial deployment for transport ticketing and payment on a local bus network in Hanau, Germany and on Taipei's Mass Transit Rail system in Taiwan. These trials have been based on the Nokia NFC shell, which clips on to a Nokia 3220 mobile phone. Future data rates up to 1.7 Mbps are currently planned, approaching the 3 Mbps of Bluetooth 2.0, and market research poi

nts to 50% of mobile phones being NFC enabled by 2010. Summary The simple PAN landscape, dominated by IrDA and Bluetooth, is becoming increasingly diverse, as shown in Figure 10-15, with many new technologies being developed that will offer the user a wide range of choices in terms of data throughput, range, power consumption and battery life. WUSB (Optional) WUSB (Mandatory) IrDA VFIR Bluetooth Bluetooth Bluetooth Class 1 Class 2 Class 3 IrDA SIR Zigbee Range (meters) Figure 10-15: PAN Technologies; Range vs. Data Rate Wireless PAN Standards Several of these technologies, such as ZigBee and NFC, have specific niche applications and in the short term the main choice for general personal area networking will be between Bluetooth and one of emerging high speed UWB radio based technologies such as wireless USB. Looking to the future, the IEEE 802.15 working group has chartered a standing committee (IEEE P802.15 SCwng) to look at the technologies that will lead to the next generation of wireless PANs. The next chapter looks at some of the considerations that impact on the choice of PAN technology for a given application, and some further practical aspects such as the security and vulnerabilities of the various PAN technologies. This page intentionally left blank CHAPTER Implementing Wireless PANs Wireless PAN Technology Choices The task of planning and implementing a wireless PAN is significantly simpler than the process discussed in Chapter 7 for wireless LANs, but the same three initial steps can also be applied here: Establish the user requirements; what is it that the user wants to be able to achieve with the PAN and what are the expectations of performance? Establish the technical requirements; what attributes does the technological solution need to possess in order to deliver these user requirements? Evaluate the available technologies; how do each of the available or emerging PAN technologies rank against the technical requirements? Establishing User Requirements User requirements are independent of specific t

echnologies and should be expressed in terms of the user experience rather than any particular solution or technical attribute. For example, in relation to battery life for mobile devices, power consumption is a technical attribute, whereas the length of time between battery recharging is what the user is really concerned about. For a PAN implementation project targeting a large user group it will be important to gather a wide range of views on user requirements - for Chapter Eleven example using a questionnaire or by interview. As a first step it may be necessary to raise awareness by demonstrating the technology to the prospective user group, SO that they are better able to give an informed view on requirements. Common types of user requirement are listed and discussed in Table 11-1. Table 11-1: PAN User Requirement Types Requirement type Considerations Usage model It is important to be clear what types of use the PAN will be put to; portable device synchronisation with a desktop or laptop computer, data exchange between portable devices, etc. Is the usage model likely to change in future or are requirements well defined and static? Device types What type of devices will be used in the PAN? Examples include laptop computer, PDA, mobile phone, hands free headset, personal video player. There may also be a requirement to connect to non-mobile devices - a desktop computer or LAN and related resources. Performance What are the user's performance expectations? This will be expectations particularly important if the usage model includes the transfer of large data files or media streaming. Device weight Particularly for PAN devices that are worn, such as a and size hands-free headset, minimising device size and weight is likely to be a requirement. Ease of connection How important is ease of connection and reconnection. If a device will be used often but intermittently, then the user may not want to perform port activation and authorisation each time a connection is made, such as is required for an IrDA link. Bluetoot

h's preferred device mode would meet this requirement. Mobility Is the network required to be moveable between locations (portability) or to operate while the user is physically moving? For example, IrDA is not suitable if any device is mobile in view of the need for port alignment. Implementing Wireless PANs Table 11-1: PAN User Requirement Types - cont'd Requirement type Considerations Device The required functionality must be implemented in both interoperability devices to enable interoperability. For example, a mobile phone may have a headset profile implemented to inter-operate with a Bluetooth headset, but will not provide Internet connectivity if the dial-up networking profile is not implemented. Operating Are there specific requirements in the environment where environment the PAN will operate, for example will it need to operate in the presence of narrow band or other RF interference, perhaps from a co-located WLAN? Battery life How often will the user need to recharge battery operated devices? Are features to conserve battery power easy to access and configure? It goes without saying that the user will want value for money in the chosen solution, but other soft issues may play a part here too. Particularly for PAN equipment, the aesthetic aspects of personal accessories may also be an implicit or explicit requirement. Establishing Technical Requirements Technical requirements follow from user requirements, translating these into the related technical attributes (Table 11-2). For example, if a user requirement is Internet access from a mobile device, then the required technical attribute is an IP networking capability. Alternatively, a user Table 11-2: PAN Technical Attributes Requirement type Considerations Application support Does the technology support the specific usage models required by the user? Effective data rate The required data rate will be dictated by the usage model, specifically the typical size of data objects that will be transmitted across the PAN, and the user's performance expectation

s in terms of upload/download time. Continued Chapter Eleven Table 11-2: PAN Technical Attributes - cont'd Requirement type Considerations Quality of service If the usage model includes applications that require isochronous data transmission, then guaranteed quality of service will be an important attribute to ensure performance expectations are met. Interference and If the PAN will have to operate in an environment with other coexistence wireless networks (e.g. an IEEE 802.11 WLAN) then coexistence will need to be a consideration. Power consumption Using PAN features on a mobile device may significantly increase power consumption, and detract from the overall performance of the device by reducing battery life. Network features, such as actively searching for other devices, should be easy to deactivate when not required in order to maximise battery life, for example on a Bluetooth enabled PDA. Operating system Operating system compatibility will be an issue when and other software applications attempt to compatibly inter-operate over the compatibility PAN link. Additional software components may be required, for example to exchange data between a mobile phone and PDA. Technology maturity Considerations will vary with the stage of maturity; before standards have been agreed, early products have an interoperability risk; a fully mature technology may have limited scope for future development and risk early obsolescence as new usage models arise. Operating range This is less important for a PAN, since by definition only a limited operating range is required - the personal operating space. However, PAN technologies vary widely in their achievable range - from 0.1m for NFC to over 100m for ZigBee. With an increasing range of capabilities coming to market in the next few years, significant price differentiation between PAN technologies can be expected. If options like ZigBee and NFC meet the user requirements they will be considerably cheaper than Bluetooth or the UWB alternatives. Implementing Wireless PANs requiremen

t to stream video to a handheld media player will translate into technical attributes for QoS and very high data rate. Evaluating Available Technologies Having established the technical attributes necessary to meet user requirements, the available technologies can then be directly assessed against these attributes. A simple table can be used to display the comparison, similar to the example shown in Table 11-3. Although the range of available PAN technologies is growing, the choice is still sufficiently narrow not to require sophisticated evaluation methods, such as assigning a relative weight to the various requirements. This approach will result in a transparent and objective comparison of the available solutions, but an independent reality check will always be helpful to verify the proposed solution - to ensure that no requirements have been missed and no limitations of a particular technology have been overlooked. It is also helpful to research the solutions that have been adopted by others to meet similar needs. If no examples can be found of the proposed solution being used in practice (IrDA for last mile broadband access?) then either a new technology application has been spotted or something has been missed in the evaluation. Table 11-3: PAN Technologies; Technical Attribute Comparison Requirement type Bluetooth ZigBee Effective data rate ca. 3 Mbps 480 Mbps 250 kbps ca. 2 Mbps 16 Mbps (2.0) (VFIR) Quality of service Interference and coexistence Power consumption Very low Very low Technology maturity Very mature mature Operating range < 10 m 10-30 m 70-300 m < 0.2 m Chapter Eleven Pilot Testing If the PAN implementation project is targeting a significant number of users then, as for WLAN implementation, a pilot testing phase will be beneficial. This will be the opportunity to confirm that the statement of user requirements is clear and complete, and that the users' performance expectations are also well defined and can be met by the proposed solution. The group of users chosen for the pilot test should co

ver the full range of technical capabilities present in the final user group - from the most technically savvy, whose performance expectations will be hardest to meet, to those who may be less demanding on performance but will have a natural focus on ease of use. Taking account of a wide diversity of views on how well the solution meets the full range of user requirements may make the implementation task more challenging, but will ultimately result in wider user acceptance of the end result. Wireless PAN Security Although a wireless PAN will generally have a more limited range than a WLAN (with the possible exception of ZigBee networks), ensuring security will remain an important implementation issue, since most wireless PAN technologies are potentially vulnerable to a variety of security threats. The following sections summarise the security features and known vulnerabilities of various PAN technologies, and provide guidance on security set-up during implementation. Bluetooth Security Bluetooth Security Overview Bluetooth includes comprehensive security measures designed to ensure that access to services is protected and only granted to another device after appropriate authorisation. Three types of service security levels are defined, as shown in Table 11-4. The first step in the process of establishing a secure Bluetooth connection is authentication, which occurs after the initial pairing and results in the Implementing Wireless PANs Table 11-4: Bluetooth Service Security Levels Security Service type Security level Open services These services can be accessed by any device. There are no security requirements and authentication and encryption are bypassed. Authentication only These services can only be accessed by services authenticated devices. Authentication and These services may only be accessed by trusted authorisation services devices. creation of a semi-permanent authentication key that is shared by the two devices. Authorisation is the second step, and may be required before a device will give another de

vice access to a requested service. Authorisation can be completed without user intervention if the requesting device is marked as "trusted". Trust is normally granted to a device by the user during an initial authorisation. Three levels of device security are also defined, as shown in Table 11-5. As a third security step, transmitted data can be encrypted using a key generated from the existing authentication key. The maximum length of the encryption key, up to 128-bit, is negotiated between master and slave Table 11-5: Bluetooth Device Security Levels Device type Security level Trusted devices Devices which are identified in the security database as trusted and are allowed unrestricted access to all services. Known untrusted Devices which have been paired and perhaps authenticated devices but which are not identified in the database as trusted. Access to certain services may be restricted. Unknown devices Devices which have not been paired and for which no security information is known. Only open services are accessible to unknown devices. Chapter Eleven as part of the process of initiating encryption. Although it is impossible to prevent the interception of data transmitted by radio, the use of FHSS makes Bluetooth virtually immune to eavesdropping by a device that does not follow the same hopping pattern. Bluetooth Vulnerabilities Provided that security modes above 1 are enabled and reasonably long passphrases or PINs are used, Bluetooth security generally prevents unauthorised access to data or services on enabled devices. However, there are two known vulnerabilities, "bluesnarfing" and "bluebugging" that affect some mobile phones, although software upgrades have been developed by vendors for phones affected by these vulnerabilities. "Bluejacking" is not strictly a security vulnerability, but represents a subversion of the normal pairing process that may lead to undesired pairing to another device. Bluesnarfing enables a hacker to access data stored on a Bluetooth mobile phone without alerting the phone's us

discoverable, in a secure environment, when necessary to establish new connections. Maximum PIN length Using the minimum 4 character PIN makes it easier for a hacker to intercept. This vulnerability can be eliminated by using PINs of at least 8 characters and preferably of the maximum allowed length. Make sure no devices are using a default PIN. Security mode Use authentication and encryption (Security Mode 3) for any confidential communications. In a multiple hop link, ensure that all links in the communication chain are using the required security mode. Anti-virus software Anti-virus software can be installed on many Bluetooth and security updates devices in the same way as on a personal computer. Anti-virus software, as well as device operating software, should be kept up to date with manufacturers' revisions and security updates. Software downloads Only download or install software from trusted sources. Careful attention should be given to any security warning during software installation. Unpair from lost If a Bluetooth device is lost or stolen, it should be unpaired devices with all devices it was previously paired with, by deleting the lost device from the list of paired devices on these devices. Failure to do this will make these devices vulnerable to attack by the previously paired device. Wireless USB Security The goal of wireless USB security is to provide the same level of confidence that the user has when making a wired USB connection, namely that the devices connected are only those that the user wants to be connected and that the transmitted data is protected from unwanted external observation or modification. Chapter Eleven Table 11-7: Fixed Symmetric Key Authentication Steps Authentication step Description Device distributes its FSK This key may be either printed on the device or included to the user in the installation software. User transfers the FSK User confirms trust of the device carrying this FSK and to the host instructs the host to allow this new connection. Host confirms to device that

By its knowledge of the device key the host is able to a new connection is allowed demonstrate to the device that it too has the user's trust. User instructs device to This may be for example by pushing a "Connect" button start new connection on the device. Authentication will include the manual entry or confirmation of a connection key (a PIN or passphrase) by the user, to ensure that hosts and devices are mutually able to demonstrate user trust when requesting or allowing a connection. Three different types of authentication "ceremony" are possible, depending on whether the connection key is distributed directly by the user to host and device, is hard wired into the device, or is based on an exchange of public keys between host and device. The authentication ceremony that most closely mirrors the process of making a wired USB connection is the second of these, based on sharing a Fixed Symmetric Key (FSK) that is typically hard wired into the device at manufacture. The steps in this authentication ceremony are shown in Table 11-7. The ceremony for public key authentication will be similar, except that both device and host will present their public keys to the user as part of a software driven installation process. After these authentication steps, association continues with a handshake process resulting in the generation of a 128-bit AES encryption key that is used to protect the connection. This pair-wise temporal key (PTK) is unique to the connection and is used by host and device to encrypt all transmitted and decrypt all received data packets. ZigBee Security The IEEE 802.15.4 MAC layer specifies four services that are available to implementers in the Zigl Bee security software toolbox to ensure security Implementing Wireless PANs Table 11-8: IEEE 802.15.4 MAC Security Services Security service Description Access control The network co-ordinator acts as a "trust centre", maintaining overall network knowledge, including a list of trusted devices within the network, as well as maintaining and distributing netw

ork keys. Data encryption Optional 128-bit AES using link keys between devices or a common network key. Frame integrity check Check to ensure that data transmitted within a frame has not been modified Sequential freshness A sequentially updated freshness value that allows the check network controller to check for and reject any replayed data frames. and data integrity, as shown in Table 11-8. Specific security implementations can then be developed using these services. Two security modes are defined in the ZigBee standard, commercial and residential. The full access control functionality of the network co-ordinator, or trust centre is only available in the commercial security mode. In residential mode the network co-ordinator controls device access to the network but does not establish or maintain keys, in order to reduce the memory cost of the trust centre device. IrDA Security The IrDA standard does not include a specification of link level security as the short range and line-of-sight requirements provide an inherent low level security. Any threat of unauthorised access to data through an active port, for example when using an IrDA enabled laptop in a public setting, can be easily countered by disabling the IrDA port when not in use and by ensuring that sensitive data is only transferred in a private environment. Additional security measures, such as authentication and encryption, are implemented at the application level. One such example is the IrDA OBEX authentication mechanism, which requires a user-entered OBEX password to be stored on both devices before an OBEX connection Chapter Eleven can be made. This password is then used by both devices for authentication when the link is established. Summary of Part IV In common with wireless LANs, wireless PAN implementation can also benefit from a systematic approach to establishing requirements and selecting the most appropriate technology. Pilot testing can also be a valuable implementation stage, particularly if the WPAN is being deployed for a large and diver

se user group. Despite their inherently shorter range, WPANs are also susceptible to a range of security threats and appropriate security measures should be considered, in line with the level of risk to user data from unauthorised access. WIRELESS MAN IMPLEMENTATION Introduction Although Wi-Fi has provided a basis for many small to medium scale metropolitan area networking initiatives, it is only with the completion of standards specifically aimed at providing wireless "last mile" solutions, such as IEEE 802.16, that the prospect has opened up of more widespread MAN applications. The key requirements for metropolitan area networking are; Scalability to hundreds or thousands of subscribers rather than perhaps tens or hundreds of users on a LAN Flexibility to provide access for a wide range of different service types, including mechanisms for requesting and allocating bandwidth Guaranteed quality of service (QoS) when required by individual subscribers or services. Despite the recent advances discussed in Chapter 6, LAN standards such as IEEE 802.11 still fall short of providing these requirements, hence the need for specific standards to address the requirements for metropolitan area networking. In Chapter 12 the IEEE 802.16 standards will be described, focussing on how these key requirements have been achieved. 802.16's European Part Five sibling HIPERMAN, which has been developed by ETSI alongside the IEEE standard, will also be briefly covered. Chapter 13 will address MAN implementation, covering the design and start-up of a wireless metropolitan area network or its rural equivalent. This will cover technical planning and implementation aspects - site surveying, equipment planning and installation - and also the business planning aspects - customer mapping, competitor analysis and management and financial planning. CHAPTER Wireless MAN Standards The 802.16 Wireless MAN Standards Origins and Main Characteristics The IEEE 802.16 set of standards have been developed since 1998 in response to the need for a wireles

s solution to supplement xDSL and cable modems in delivering broadband access to homes and small businesses. A wireless solution for "last mile" broadband access has the advantage of being able to provide wide geographical coverage with minimal infrastructure cost, and therefore brings with it the potential to accelerate broadband uptake. The evolving suite of IEEE 802.16 standards is shown in Table 12-1. The 10-66 GHz frequency range was the initial focus of the IEEE Working Group on broadband wireless access (BWA) which developed the standards. This was largely motivated by worldwide spectrum availability in this frequency range, and led to the approval of the initial IEEE 802.16 standard in 2001 and its publication in March 2002. The Working Group then turned its attention to the 2-11 GHz range, where the advantages of lower cost implementation and non line-of-sight transmission outweighed the potential difficulties in this crowded piece of RF spectrum. The result was the IEEE 802.16a standard, approved in 2002 and published in January 2003. IEEE 802.16 has been designed to offer considerable flexibility in the specification of the PHY layer, in order to accommodate varying requirements (such as channel widths) in different regulatory regimes. Chapter Twelve Table 12-1: The IEEE 802.16 Standard Suite Standard Key features 802.16 Original standard, approved in 2001. 10-66 GHz spectrum, line-of-sight links up to 134 Mbps. 802.16a Approved in 2002. 2-11 GHz. Non line-of-sight links up to 70 Mbps. 802.16b Update to 802.16a dealing with unlicensed applications in the 5 GHz band. 802.16c Update to 802.16 addressing interoperability of 10-66 GHz systems. 802.16d The basis of "WiMAX", replaces 802.16a and includes support for advanced antenna systems (MIMO). Approved in June 2004 as 802.16-2004. 802.16e Extension to provide mobility, including rapid adaptation to the changing propagation environment. 802.16f Extension to support multi-hop capabilities required for mesh networking. 802.16g Addresses efficient handover

and QoS for mobile networking. 802.16h MAC enhancements to enable coexistence of licence exempt 802.16 based systems and primary users in licensed bands. These different air interfaces are supported by a common MAC layer, which has been designed to provide the key requirements for metropolitan area networking - scalability, service type flexibility and quality of service. 802.16 PHY Layer PHY Layer for 10-66 GHz Spectrum In this extremely high frequency range, RF propagation requires for all practical purposes a line-of-sight between transmitter and receiver. Given this limitation, it is not necessary to consider the use of complex techniques such as OFDM to overcome multi-path effects, which occur in a non-line- of-site environment, and the Working Group therefore selected a simple single carrier (SC) modulation technique for this interface (Table 12-2). In downlink transmissions (from the base station (BS) to subscriber stations (SS)), time division multiplexing (TDM) is used, with time slots Wireless MAN Standards Table 12-2: 802.16 Key Parameters Parameter 802.16 standard RF band 10-66 GHz Modulation Single carrier modulation (SC) (QPSK, 16- & 64-QAM) Data rate Peak data rates to 134 Mbps Channelisation 20, 25 or 28 MHz channel widths Duplex method TDD and FDD, as well as half-duplex via TDMA Network topology Point-to-multipoint Bandwidth allocation Grant per subscriber station (GPSS) - see the Section "802.11 PHY Layer, p.310" allocated to individual subscribers. This enables bandwidth to be guaranteed to latency sensitive services. In the uplink direction (from SS to BS) time division multiple access (TDMA) is used. Duplexing of up and downlinks can be achieved either by time division or frequency division duplexing (TDD or FDD). Half-duplex subscribers, transmitting and receiving on the same channel, are also supported by TDM portion Broadcast TDMA portion control (DIUC 1) (DIUC 2) (DIUC n-1) Downlink Uplink (DIUC n) (DIUC n+1) (DIUC n+2) (DIUC m) Downlink Interval Usage Codes (DIUC), in the Downlink Burst

profile Map, specify the burst profile to each subscriber station start points Figure 12-1: 802.16 Downlink Frame Structure Chapter Twelve having an optional TDMA uplink segment following the TDM portion of each downlink data frame. A range of modulation and coding schemes are available (including QPSK, 16-QAM, 64-QAM) and each subscriber station negotiates a scheme with the base station, in line with its particular needs for efficiency (depending on data rate) and robustness (depending on signal propagation environment/signal strength). The result is that, within a single downlink frame transmitted by the base station (Figure 12-1), individual data bursts destined to individual subscriber stations will be transmitted with different coding and modulation schemes - a technique known as adaptive burst profiling. Achievable data rates for different modulation methods are shown in Table 12-3. Table 12-3: 802.16 Bit Rate vs. Channel Width and Modulation Method Channel width Bit rate (Mbps) per modulation method (MHz) 16-QAM 64-QAM 134.4 PHY Layer for 2-11 GHz Spectrum The differing propagation characteristics in the 2-11 GHz range, compared to the EHF range up to 66 GHz, require an air interface that can accommodate significant multi-path propagation in a non line-of-site operating environment. Three optional PHY layer specifications are defined in the 802.16a standard, which addresses both licensed and unlicensed spectrum, as shown in Table 12-4. Most of the key parameters of the 802. 16a standard are common to its higher frequency predecessor, as shown in the Table 12-5. In addition, 802.16a also supports advanced antenna systems. As described in the Section "Wireless LAN Antennas, p. 55", advanced antenna systems or smart antennas can improve link robustness by suppressing interference and increasing overall system gain. As the cost Wireless MAN Standards Table 12-4: 802.16a Optional Air Interfaces Air interface Summary WirelessMAN SC2 A single carrier modulation format, providing interoperability with the 10-66 G

Hz single carrier air interface. WirelessMAN OFDM Orthogonal frequency division multiplexing, with TDMA controlled multiple user access. This interface is mandatory for unlicensed bands. WirelessMAN OFDMA An OFDM interface with user access controlled by allocating to individual users a subset of the full set of available carrier frequencies. of these systems comes down, they will play an important role in improving wireless network performance, particularly in the increasingly crowded unlicensed frequency bands. MAC Layer In fulfilling the needs of metropolitan area networking, the 802.16 MAC layer must provide flexible and efficient access for a wide range of different service types. The main shortcoming of the contention based Table 12-5: 802.16a Key Parameters Parameter 802.16a standard RF band 2-11 GHz Modulation Single carrier, OFDM Data rate Peak data rates to 70 Mbps Multiple access OFDMA, TDMA Channelisation Flexible channel widths, from 1.75 to 20 MHz Duplex method TDD and FDD, as well as half-duplex via TDMA Network topology Point-to-multipoint and mesh topologies Bandwidth allocation Grant per subscriber station (GPSS) or grant per connection (GPC) - see Section 12.1.3 "MAC Layer" Chapter Twelve media access of 802.11 networks, based on carrier sensing (CSMA-CA), is that, prior to the 802.11e enhancements, no particular service quality level could be guaranteed. A subscriber requiring a latency-sensitive service such as VoIP may be subject to the exposed station or hidden station problems (Section "Radio Transmission as a Network Medium, p. 75") with the consequent deterioration of service quality. Connection-Oriented Versus Connectionless One key to the effectiveness of the 802.16 MAC is its connection- orientation. Every service is mapped to a connection and referenced using a 16-bit connection ID (CID). This includes services which are inherently connectionless such as UDP services (for example RIP, SNMP or DHCP messaging). Each connection can then be associated with specific parameters such as; ban

dwidth granting mechanism (continuous or on-demand) Associated QoS parameters Routing and transport data. Connections are typically unidirectional, SO that different QoS and other transport parameters can be defined for the uplink and downlink directions. When a new subscriber station joins an 802.16 network, three connections are initially opened to carry management level messages, as shown in Table 12-6. Table 12-6: 802.16 SS Management Connections Connection Usage Basic management connection Short, time-critical MAC and radio link control (RLC) messages. Primary management connection Delay tolerant messages e.g. authentication, connection set-up. Secondary management Standards based management messaging; connection DHCP, SNMP, RIP. Wireless MAN Standards Additional connections are allocated to subscribers when specific services are contracted, typically in uplink plus downlink pairs. The MAC also reserves a number of connections for other general purposes, such as initial contention-based access and broadcast or multicast transmissions, including polling of SS bandwidth needs. Radio Link Control Radio link control (RLC) is another key element of the 802.16 MAC that is required to provide adaptive burst control as well as the more traditional power adjustment functions (TPC). When a subscriber joins the network, SS and BS exchange messages using the basic management connection to establish initial settings for transmit power and timing. The SS will also request a specific initial burst profile, defining signal modulation parameters, based on equipment capabilities and downlink signal quality. The RLC will continue to monitor signal quality after this initial set-up, and either the SS or BS may request a more robust burst profile if environmental conditions deteriorate (e.g. temporarily switching from 64-QAM to 16-QAM) or a more efficient profile if lower robustness can be tolerated as conditions improve. The uplink burst profile is under the direct control of the BS, and this control is achieved each time the B

S allocates bandwidth to a SS by also specifying the burst profile to be used by the SS. The downlink profile is also controlled by the BS, although changes are at the request of the SS, which alone can monitor the strength of the received signal. Bandwidth Allocation Flexible allocation of available bandwidth among subscribers and connections is a third key element of the 802.16 MAC. The variety of services and scalability to potentially hundreds of subscribers per BS will clearly put heavy demands on the efficient use of bandwidth. The bandwidth requirements of individual subscribers are established when a connection is made and a number of messaging options in the standard allow SSs to request additional uplink bandwidth, to inform the BS of total bandwidth needs or allow the BS to poll individual SSs or multicast groups to establish these requirements. These mechanisms ensure efficient use of available bandwidth and flexibility to Chapter Twelve Table 12-7: 802.16 GPC and GPSS Classes of Subscriber Station SS class Capabilities GPC subscriber station Bandwidth granted to the SS can only be used for the connection that requested it. GPSS subscriber station Bandwidth granted by the BS need not be used only for the connection requesting it. The SS can use granted bandwidth for any of its connections. accommodate a wide variety of services or to respond to changing requirements of individual services. The MAC defines two classes of SSs (Table 12-7) - Grant per connection (GPC) and grant per subscriber station (GPSS), which differ in terms of the flexibility available to the SS in using its allocated bandwidth. At the cost of some additional complexity, GPSS provides greater efficiency and scalability than GPC, for example allowing a SS to respond more quickly to changing environmental conditions. GPSS requires additional intelligence in the SS in order to manage the QoS of its connections, but is clearly one aspect of the autonomy that has to be delegated to subscriber stations in a mesh network. Mobile WiMAX The

802.16 Task Group TGe addressed the challenge of adding mobility to the 802.16 standard. A study group on Mobile Broadband Wireless Access was set up in 2002 with the aim of providing mobile access at vehicular speeds of up to 125 km/hr using licensed frequency bands. The resulting 802.16e standard, also designated 802.16-2005, was ratified by the IEEE in December 2005 and specifies new modulation and multiple access schemes to enable mobile non line-of- sight operation. 802. 16e enhances the 802.16's original OFDMA air interface by adding a number of new capabilities, as summarised in Table 12-8. The two key concepts underlying the 802.16e enhancements are the fixing of subcarrier spacing independently from channel width (scalable OFDMA) and the use of subchannelization to enable range versus capacity trade-offs. Wireless MAN Standards Table 12-8: 802.16e Enhancements Enhancement Capabilities Constant subcarrier spacing Increased resistance to multipath fading and Doppler spread in mobile transmission is achieved by keeping subcarrier spacing constant, independent of channel width. This results in a "scaling" of the number of subcarriers with channel width. Improved indoor penetration A subset of the available OFDM carriers is used at higher individual power (subchannelisation) to improve indoor reception. Flexibility in coverage versus Subchannelisation in the downlink allows a flexible capacity trade-off between data capacity and operating range. Advanced antenna support NLOS coverage and performance are improved using advanced multi-antenna diversity and adaptive antenna systems as well as MIMO radio technology. Enhanced error correction New coding techniques are used to improve mobile and NLOS performance (e.g. turbo coding and low-density parity check (LDPC)). Faster error recovery Using hybrid-automatic retransmission request (hARQ) to improve error recovery. Scalable OFDMA As described earlier, 802.16a specifies an OFDM air interface with flexible channel widths from 1.75 to 20 MHz. Each channel is divid

ed into 256 subcarriers (OFDMA256), SO that the subcarrier spacing depends on the channel width and varies from 6.8 to 78.1 kHz. In mobile applications, varying Doppler shift and multipath delays lead to a degradation of performance, in terms of SNR and BER, particularly for subcarrier spacings at the low end of this range. Conversely, channel capacity can be increased by using more subcarriers at higher channel widths. Scalable OFDMA addresses these issues by using a constant subcarrier spacing of 11.2 kHz and varying the number of subcarriers depending on the channel width, from 128 for a 1.25 MHz channel to 2048 for a 20 MHz channel. This maximises channel capacity for the wider channels and ensures that all channel widths are equally tolerant of the delay spread resulting from moving stations. Chapter Twelve Full channel of OFDM carriers Uplink subcarrier power if all tones are used Pilot Subcarrier power tones for adjacent subchannelisation Carrier power for distributed subchannelisation Figure 12-2: WiMax Subchannelization S-OFDMA and OFDMA256 will not be compatible, SO that WMAN equipment will have to be replaced in order to support mobile applications based on S-OFDMA. Subchannelization Subchannelization refers to the use of a subset of available OFDM subcarriers. By concentrating the available transmitter power into fewer subcarriers, each subcarrier can be transmitted with higher individual power. This additional link margin can be used either to extend the link range, to overcome penetration losses allowing mobile devices to be located indoors, or to reduce the power consumed by the transmitting device. These benefits come at a cost of reduced link capacity, since only a subset of subcarriers are used to carry data, allowing a trade-off between throughput and mobility. Various methods can be used for allocating subcarriers to subchannels - the main ones being the use of adjacent or distributed subcarriers (Figure 12-2). Distributed subcarrier allocation is used for mobile applications mainly because th

e use of a wide range of frequencies (frequency diversity) makes the link less susceptible to the rapidly varying fading that is characteristic of mobile applications. Subchannelization is also an option in fixed WiMax, where a subchannelized uplink can be used to trade-off throughput for range, achieving a higher range or increased building penetration for a given CPE transmitter power. Wireless MAN Standards IEEE 802.16 in Practice The familiar model of an IEEE standard being taken up by a trade organisation to drive product and market development is also applicable for 802.16. The WiMAX (Worldwide Interoperability for Microwave Access) Forum aims to fulfil a similar role to that of the Wi-Fi Alliance in relation to the 802.11 suite of standards, namely conformance and interoperability testing and certification. The WiMAX Forum is marketing 802.16d networks under the WiMAX brand as providing fixed, portable and ultimately mobile wireless broadband access without requiring direct line-of-sight to a base station. Data capacities of up to 40 Mbps per channel are headlined, sufficient to support thousands of residential subscribers at DSL connection speeds, while mobile networks are expected to reach 15 Mbps. Initial certification work by the WiMAX Forum members will focus on equipment operating in the 3.5 GHz frequency band with 3.5 MHz channel width, and based on both TDD and FDD. Extensions beyond this initial scope will be dependent on further market demand and product submissions by vendors. Future Developments Further developments of the 802.16 standards suite are currently being progressed by the TGf and TGg Task Groups. 802.16f aims to improve the multi-hop capabilities, as SSs move between BSs in a fixed wireless network, while 802.16g will deliver faster and more efficient handover and improved QoS for mobile connections. Other WMAN Standards ETSI HIPERMAN HIPERMAN, which stands for high performance radio metropolitan area networks, is a proposal from the European Telecommunications Standards Institute (E

TSI) to provide "last mile" fixed wireless access to the small and medium enterprise and residential markets, with the primary aim of accelerating the uptake of broadband Internet access within Europe. The standard has been developed in close co-operation with the IEEE 802.16 Working Group, and is intended to be interoperable with a subset of the 802. 16a standard - namely the OFDM air interface described in Chapter Twelve Table 12-9: HIPERMAN Key Parameters Parameter HIPERMAN standard RF band 5.725-5.875 GHz (European Band C) < 30 dBm (1 watt) Data rate Peak data rates > 2Mbps Modulation OFDM (BPSK, QPSK, QAM) Channelisation 5 MHz, 10 MHz or 20 MHz channel widths Duplex method TDD and FDD are supported the Section "802.11 PHY Layer, p. 310". Table 12-9 shows the key parameters of the HIPERMAN standard. ETSI's intent is to develop a standard which uses unlicensed spectrum, but which can support the commercial provision of fixed wireless access (FWA) based on either point-to-multipoint (PMP) or mesh network configurations. The channel definition is intended to enable at least two competing service providers to cover a metropolitan area, more if the narrower (5 MHz or 10 MHz) channels are used. HIPERMAN defines only the PHY and Link Control layers and, as for 802.16d and the WiMAX Forum, it is anticipated that specifications for the Network layer and higher layers will be developed by other bodies. The equivalent "HIPERMAN Forum" is yet to emerge, but the WiMAX Forum has committed to addressing the issue of interoperability, SO that HIPERMAN is incorporated into the WiMAX standard. The WiMAX forum will effectively fulfil the commercialisation role for both wireless MAN standards. TTA WiBro WiBro, short for Wireless Broadband, is a wireless MAN standard developed by the Telecommunications Technology Association (TTA) of South Korea, Phase 1 of which was approved in November 2004. The standard was developed to fill the gap between 3G and WLAN standards, providing the data rate, mobility and coverage required to deliv

er Internet access to mobile clients via handheld devices. The standard uses 100 MHz of licensed RF spectrum, from 2.30 to 2.40 GHz, allocated by the South Korean Ministry of Information Wireless MAN Standards Table 12-10: WiBro Key PHY Layer Parameters Parameter WiBro standard RF band 2.300-2.400 GHz Network topology Cellular structure, ca. 1 km range Maximum data rate; User Downlink; 6 Mbps, uplink; 1 Mbps Maximum data rate; Cell Downlink; 18.4 Mbps, uplink; 6.1 Mbps Multiple access OFDMA Modulation QPSK, 16-QAM, 64-QAM Channelisation 9 MHz channel widths Duplex method and Communication for mobile wireless Internet usage, and adjacent to the international unlicensed 2.4 GHz ISM band. The IEEE 802.16-2004 and Draft 3 of the 802. 16e standard were the basis for the development of WiBro, and the key PHY layer parameters, shown in Table 12-10, are compatible between the two standards. The WiBro MAC supports three service levels including guaranteed QoS for delay sensitive applications, based on real time polling of station requirements, and an intermediate QoS level for delay tolerant application that require a minimum guaranteed data rate. Phase 2 of the standard is planned to focus on network capacity enhancement technology, including MIMO radio, adaptive antenna systems and space time coding, as well as further standardisation with 802.16e-WiMAX Metropolitan Area Mesh Networks The wireless MAN standards described above, enable point-to-point or point-to-multipoint solutions for metropolitan area networking, with dedicated base stations providing centralised control. Although no metropolitan area mesh networking standard is currently under development, the work of 802.11 Task Group TGs described in the Section "Mesh networking (802.11s), p. 167" will blur the boundary between Chapter Twelve LAN and MAN scales, allowing 802.11 based mesh networks to operate effectively over metropolitan areas. A mesh based approach will have a number of advantages over the traditional MAN topologies, including making optimum use o

f alternative paths through the mesh to maximise network throughput and automatically taking advantage of any new backhaul links that become active in the mesh operating area. Proprietary (i.e. non standards based) equipment is currently available that operates a pseudo 802.11b mesh, with fixed "mesh routers" providing metropolitan scale coverage for mobile 802.11b devices (see the Section "Wireless LAN Resources by Standard, p. 367"). Summary Although the WiMAX Forum members are yet to bring IEEE 802.16 compliant products to the marketplace, it looks set to become the de facto standard in this sector of the wireless networking market. The 802.16 MAC enables bandwidth allocation and quality of service to be specified for individual connections, provides adaptive burst profiling to allow the most efficient modulation and coding methods to be used according to each subscriber station's capabilities and environmental conditions, and provides a variety of mechanisms to vary bandwidth allocation according to the requests of individual subscriber stations. Through these features, the 802.16 standard meets the scalability, flexibility and quality of service requirements for metropolitan area networking. The rapid development of the WiBro standard by the South Korean consumer electronics industry is the first visible chink in the IEEE's armour. Targeted at a specific niche-need in the highly "net-savvy" South Korean consumer market, it shows that speed of development of standards is one area where the IEEE process can fall short of market needs. Nevertheless, WiBro remains based on the 802.16 specifications, stressing the importance of standardisation in global markets. Chapter 13 now turns to the practical aspects of planning and implementing wireless MANs. CHAPTER Implementing Wireless MANs When setting out on the implementation of a wireless MAN project, the starting point will be a specific area envisaged as the general target area of the network. This may be an urban area such as a town or city centre, where potenti

al subscribers will be relatively concentrated, or a more distributed rural area. In the first case the wireless MAN is more likely to be in competition with other broadband access options, such as cable or xDSL, while, in the rural setting, wireless may be the only "last mile" access option available to prospective subscribers. Two aspects of planning need to be addressed in implementing wireless MANs - technical planning and business planning. The first looks at what is required to build an effective MAN from the technical standpoint - the physical hardware, its specification, location, installation and operation, while the second looks at what is required to operate the MAN as a successful and profitable business. The most important factors to ensure success in this area are an understanding of the market, insight into the competitive situation, and well thought out business and financial plans. Technical Planning Technical planning starts with an understanding of the customer demographics in the intended MAN operating area. Based on this understanding, together with a physical and RF survey of the target area, an equipment plan can be developed to achieve maximum effective coverage at minimum capital cost. Future growth of the MAN will be eased by also considering a cost-effective migration path in the planning at this stage - even if this may seem premature. Chapter Thirteen Site Surveying Site surveying for a wireless MAN aims to assess the physical and RF environment in which potential subscriber stations will have to operate, for example in terms of physical obstacles and potential sources of interference. A general site survey should be conducted over the whole target area at the start of the technical planning of the network, to assess the main constraints and consideration that will impact on the overall network design. Later, in the start-up and operating phase, specific surveys of subscriber sites will often be required in order to ensure that the required quality of service can be delivered before c

ommencing physical installation. A variety of simulation programs, often based on the AT&T Wireless model and more recently the Stanford University Interim (SUI) models, can be found online that can assist in the survey process (see the Section "Wireless MAN Resources by Standard, p. 369". These tools can be used to provide an initial coverage estimate, but they should be used with caution as the performance of the network will be largely determined by local environmental conditions that can only be properly accounted for by on-site physical and RF surveys. The results of the physical and RF surveys will provide essential input to network planning, determining the best network equipment specification and configuration to avoid potential sources of interference as well as line- of-sight obstacles, while allowing for other specific local conditions. As discussed below, antenna selection is perhaps the most important element in equipment planning, and the suitability of the selected base station and subscriber antennas will be a key factor in determining the ultimate performance of the network. This selection will be based in part on the site survey results. Physical Site Survey Considerations A physical site survey will check the visual line-of-sight from various points in the target area to the potential base station locations. Fresnel zone clearance (see the Section "Fresnel Zone Theory, p. 113") should also be considered as well as the direct line-of-sight (Figure 13-1). The survey should consider local topography, possible obstructions such as tall buildings, and the proximity to locations such as airports where radar may be a potential cause of interference (this will be checked in the RF survey). Implementing Wireless MANs Business district; LOS Existing towers poor between buildings Fresnel zone problems around business district Low ground elevation; poor LOS from east and west Leafy suburbs; seasonal foliage cover Figure 13-1: A Physical Site-Survey Map for a MAN Installation At the subscriber set-up stage,

the physical site survey can include specific customer premises equipment (CPE) siting, as well as cable routing and other needs such as lightning protection in exposed locations. RF Site Survey This part of the site survey will assess the RF environment in which subscriber stations will have to operate in terms of potential sources of noise and interference. With this objective in mind, the survey should ideally be conducted with an antenna similar to the CPE equipment that will be installed on the subscriber's premises. An RF site survey is conducted using a spectrum analyser, which will identify wireless transmissions in the target area and in the frequency range of interest. An example of a spectrum analyser display, showing vector analysis of an OFDM signal, is shown in Figure 13-2. The signal strength, direction and polarisation of any signals strong enough to interfere with the network should be recorded, as well as the noise floor (see the Section "Receiver Noise Floor, p. 110"). Spectrum analysis software is also available to run on a desktop or laptop computer and will analyse the signal received from a PCI or PC card receiver equipped with an external antenna. Chapter Thirteen Agilent 89600 Vector Signal Analyzer Control Source Input TestSetup MeasSetup Display Trace Markers Utilities Grid 2x2 Color Normal A: Strm1 OFDM Meas Strm2 OFDM Meas Const Const -1.25 -1.25 -1.722 1.7225 -1.722 1.7225 RBW: 312.5 L kHz TimeLen 8 Sym RBW: 312.5kHz TimeLen 8 Sym B: Strm1 OFDM Eir Vect Spectrum D: OFDM Data Burst Info ModF Fmt Len(sym) Pwr(dBm) EVM(dB) L-STF -5.5788 HT-LTF -5.5601 HT-SIG -5.8612 LinMag HT-LTF -5.4579 HT Data 64QAM -5.5901 -39.685 Total -5.6107 -39.685 HT-SIG CRCPassed carrier Stop 58 carrier RBW: 312.5kHz TimeLen 8 Sym Figure 13-2: Spectrum Analysis Software Display From Agilent (Courtesy of Agilent Technologies Inc.) The results of the RF site survey will be important considerations in the technical design of the MAN, and may impact on aspects such as frequency band selection and base station equi

pment specifications and location. Capital Equipment Selection and Location Having completed both physical and RF site surveys, and established potential base station locations to serve the target area, a link budget can be calculated for typical and extreme subscriber locations within the target operating area. This will establish the transmitter power and antenna gain requirements at each base station as well as the CPE antenna gain required to achieve the desired system performance, taking account of any regulatory limitations (such as EIRP) on the type of equipment that can be used. Four key elements need to be considered in selecting and locating capital equipment; Base station transmitter/receiver Base station antenna Implementing Wireless MANs Base station antenna location Customer premises equipment. Selection of a Base Station Transmitter/Receiver The base station transmitter/receiver (Figure 13-3) is key to achieving maximum effective coverage of the target area. Besides transmitter power and receiver sensitivity, quality and reliability are the factors which will determine overall system performance and, by keeping maintenance and downtime to a minimum, will also reduce operating costs and assure subscriber satisfaction. Base Station Antenna Selection Base station antennas come in many shapes and sizes but in most wireless MAN applications omnidirectional coverage will be required. This may be achieved with either omnidirectional or multiple sector antennas (Figure 13-4). Directional antennas may also be required, for example to extend a MAN using a point-to-point bridge. In general, within local regulatory limitations, it is preferable to provide the maximum possible antenna gain at the base station, since, for a given link budget, this allows a reduction in the specification and cost of CPE. Installing higher cost antennas at a few base stations is clearly preferred Figure 13-3: A Base Station Transmitter (Courtesy of Aperto Networks Inc.) Chapter Thirteen Figure 13-4: A Base Station Sector Antenna A

rray (Courtesy of European Antennas Ltd.) from an economic standpoint to providing the same increment in gain at many more subscriber locations. Base Station Antenna Location Having mapped out the intended area for the MAN, the local terrain characteristics and availability of suitable antenna locations would have been assessed as part of the physical site survey. The results of this survey will be analysed to define the minimum number and optimal location of base stations required to serve the target area. There are two basic options to consider in deciding how to achieve coverage of the target area - coverage from the periphery (Figure 13-5) or from the centre as shown in Table 13-1. If IEEE 802.16 is the MAN technology being deployed, an individual base station range of 3-10 kilometres should be considered when mapping Implementing Wireless MANs Table 13-1: MAN Target Area Coverage Options Option Considerations Coverage from the A number of locations are identified to provide coverage periphery of the target from the periphery of the target area. This option may be advantageous where local terrain, such as nearby hills, can provide line-of-sight coverage of a wide area. This may also allow subscribers to select the best base station from several different directions. Coverage from the In this option one or more base stations are placed on centre of the target area tall structures located within, and preferably close to the centre of, the target area. out coverage. Depending on the size and shape of the target area, a combination of peripheral and central coverage may be required (Figure 13-6). Either of these options can be supplemented by the use of additional bridge linked base stations that can be positioned to fill in parts of the target area that may be shielded by high buildings, local terrain or other obstacles. Having established the general layout of base stations required to cover the target area, Table 13-2 summarises issues that need to be considered in determining specific locations for BS antenna

s. Narrowbeam peripheral base station Widebeam peripheral base stations Figure 13-5: Peripheral Coverage of a MAN Target Area Chapter Thirteen Three peripheral Two central base stations base stations Figure 13-6: Mixed Central and Peripheral Coverage of a MAN Target Area Height is generally the greatest asset when locating BS antennas, although care must be taken to ensure that the antenna's beam pattern is directed to provide coverage close to the base of the tower or building, for example by rotating the axis of the antenna downwards below the horizon. Table 13-2: MAN BS Antenna Location Considerations Issue Impact Existing buildings or Existing buildings or towers provide a potentially towers low cost solution for BS antenna location. Barter deals (trading broadband access for antenna siting rights) can also reduce the cost of leasing antenna mounting space on a tower. New towers New towers allow optimum location - improved coverage may offset the additional cost compared to leasing space on existing structures. Local planning regulations will need to be considered in siting new towers. Local terrain Favourable ground elevation can help areal coverage by adding effective height to an antenna location - whether on the periphery or central to the target area. Implementing Wireless MANs Customer Premises Equipment A typical customer premises equipment (CPE) installation will comprise an antenna and radio enclosed in a weatherproof sealed unit, typically mounted on a suitably facing wall of the premises or on a chimney breast (Figure 13-7). The antenna may range from a patch antenna, with 8-14 dB gain for short range links, to a high gain parabolic reflector antenna with 24 dB gain and upwards for longer range. Transmitter power will depend on local regulatory limits with 100-200 mW being typical. Cabling will require outdoor rated Cat 5 Ethernet cable as well as a low voltage DC power line routed back inside the premises. Two approaches can be taken in selecting CPE - one-size-fits-all or subscriber customisation

subscriber Subscriber satisfaction At risk when the limits of Likely to be higher as a standard CPE performance result of higher level of are reached perceived service through customised installation Chapter Thirteen aperto Figure 13-7: A Typical CPE Installation from Aperto (Courtesy of Aperto Networks Inc.) The cost of installing CPE can be a significant part of the total set-up cost for each new subscriber, and ease of installation is therefore an important consideration in deciding what equipment to use. Backhaul Provision Backhaul facilities (Figure 13-8) will provide the link that connects the network base station through to an Internet gateway - the first onward destination of subscriber traffic. This link will prove easiest to achieve if there are local ISPs or other Internet points-of-presence (POPs) nearby. If a local POP is not available it will be necessary to consider local leased options such as cable or fibre optic providers. In the absence of existing backhaul infrastructure, which may be the situation in remote rural locations, wireless links such as long range point-to-point or satellite options can be investigated. Business Planning While technical planning is the key to the physical performance of a MAN, business planning is the key to turning that technical success into a Implementing Wireless MANs Option a) Point-to-point wireless link to remote POP Option b) Ethernet link to local Internet PoP Figure 13-8: Backhaul Configurations financial success. Although creating a business plan may seem to be a time-consuming effort, that effort will be quickly rewarded by helping to identify any shortcomings that can be more easily, and generally more cheaply, overcome at an early stage in the venture. The four key elements of the business plan are; A description of the business A marketing plan A management and operations plan and A financial plan. Business Description This section of the business plan will provide a simple and clear description of the planned business, setting out its purpose and the

nature of the service being offered - in this case broadband wireless access. A brief description of the target market, and the specific needs of the market that the service will address, will provide an introduction to the marketing plan. Marketing Plan The marketing plan starts with competitor analysis and customer mapping. The competitor analysis will identify alternative broadband access options that are available in the target area, such as cable, xDSL or Chapter Thirteen competing wireless providers. Some specific aspects of the competition that will need to be assessed are; What types of services are being offered by competitors? What is the market perception of the quality of service provided? What is the range of available uplink and downlink data rates? What are the initial set-up and equipment charges made by different competitors? What are the typical subscriber charges, including variable rates based on time or usage? Are promotional rates or discounts being offered? Customer mapping establishes the density of potential subscribers in the target area of the network, including the mix of business and residential customers. This will also be an input to the initial technical planning to generate a picture of the physical network deployment required to reach the maximum number of potential subscribers. To turn the customer mapping into a subscriber forecast (Figure 13-9) it will be necessary to do some market research in order to estimate an uptake rate within the target customer base. An initial pricing assumption will be needed as input to this market research. This may be based on information from the competitor analysis or on the cost of similar services in other areas. A useful approach may be to Rapid uptake High latent demand No alternative Intermediate uptake Moderate competition Slow uptake Many alternative Competitive prices and services Years of operation Figure 13-9: Example of Uptake Curves for Various Assumptions Implementing Wireless MANs test the likely customer uptake against a range o

f possible prices as this will allow a range of scenarios to be developed at the financial planning stage. However, price is not the only way to differentiate a new service from the competition. The overall service bundle can be an important differentiator in the eyes of the subscriber, SO it is important to consider creative ways to add distinctive value through additional services that can be delivered at low cost to the service provider. Examples may be web hosting, anti-spam or anti-virus screening or VoIP services. In considering these aspects of the marketing plan it is important to understand; What subscriber needs does the service fulfil and what related needs might be fulfilled? What is unique about the service being offering? Bundled services? Lowest cost? What is the intended subscriber view of the service? Low cost? Premium service? How do competitors position themselves in relation to these service and price aspects? The final aspect of the marketing plan is advertising - what approaches will be used and what will the cost of advertising be, again as input to the financial plan. Management and Operations Plan The management plan will set out how the venture will be managed; who will form the management team, what are their specific qualifications, skills and relevant experience. An assessment of the strengths and weaknesses of those involved in setting up the venture is important here; What specific expertise and qualifications do they bring to the venture? Where will the technical, business and operational skills needed to run the venture come from? Recognising areas where skills need to be supplemented, is a good starting point for selecting partners and staff. The operations plan will either be included here or as a separate section. This will cover the additional personnel requirements in the Chapter Thirteen operating phase, as well as outline operating procedures that will need to be developed later. Other issues such as business insurance can also be addressed in this section of the plan. Fina

ncial Plan The financial plan should start with a statement of the financial objectives of the venture, in terms of a target revenue or net profit level. The technical plan, generated in parallel with the marketing plan, will allow capital equipment lists and the resulting start-up budget to be prepared. Similarly monthly and annual operating budgets can be generated from supplies lists, manpower costs, etc. (Table 13-4). The results of the marketing plan and operating budget can be used to develop a number of cash flow projections, taking account of capital Table 13-4: Operating Cost Elements and Assumptions Operating cost item Business case assumption Comments Base station lease costs Negotiated costs per Covers the space required month per base station for equipment located indoors as well as antenna site lease cost Equipment surveillance Percentage (typically 5%) Maintenance and costs and maintenance of BS equipment cost, or will be higher for 7% of CPE cost if owned equipment which is by the network operator remotely located Network operations Percentage (typically 10%) Initial higher % reflects of gross revenue in first fixed start-up costs, later year, dropping (to 5-7%) years reflect stable in later years business position Sales and marketing Percentage (typically As for network operations costs (including 20%) of gross revenue customer support) in first year, dropping (to around 10%) after five years General & administrative Percentage (typically As for network operations costs 5-6%) of gross revenue in first year, dropping (to around 3%) after five years Implementing Wireless MANs Subscribers (Uptake curve b) 105.0 140.0 Monthly subscription Gross revenue 150.0 300.0 500.0 750.0 1050.0 1400.0 Base station leases 150.0 300.0 450.0 600.0 600.0 600.0 Equipment surveillance and maintenance 100.0 150.0 200.0 200.0 200.0 Network operations 105.0 140.0 Sales and marketing 100.0 100.0 100.0 150.0 210.0 280.0 General and administration Total operating costs 380.0 580.0 780.0 1070.0 1178.0 1304.0 Net cash flow -2

30.0 -280.0 -280.0 -320.0 -128.0 Figure 13-10: A 6-Month Initial Cash Flow Projection for a Start-up WISP Venture spending, direct and indirect expenses and revenues. These should be made on a monthly basis for the start-up period of one to two years and then on a yearly basis. An example start-up cash flow projection is shown in Figure 13-10. Generating multiple scenarios based on alternative uptake rates and pricing policies will illustrate the range of potential cash flow outcomes and help to assess how exposed the venture is to risk. Cash flows can be used to identify the total funding required until the venture starts to be positive cash generating. Cash flows can then be turned into income projections (profit & loss statements) and various break-even and profitability analyses can be performed (Figure 13-11). Again, multiple scenarios on customer uptake, pricing, future market development and other variables can be played out to give stakeholders an idea of the robustness of the proposed venture. Spreadsheets are an ideal tool for conducting this analysis, although custom made financial and general business planning software is also available that can guide the financial planning process. Start-up Phase Some of the key considerations that will require attention in the start-up phase are covered in this section. Chapter Thirteen 15000 Monthly cash flow 12500 Cumulative cash flow Breakeven from 10000 operating month 9 Cashflow positive from operating month 6 Maximum cash deficit in operating month 5 -2500 MJJASONDJFMAMJJASOND Figure 13-11: A Graphical Break-even Analysis Chart Base Station Deployment Tower leasing - Where existing towers are the preferred option, contracts will have to be negotiated to acquire space to locate BS antennas. Sample tower leasing agreements can be found on the Internet, although it is recommended to take legal advice to ensure that agreements comply with local laws. New tower deployment - When selecting a tower design, current needs and requirements for future expansion should be

considered. Construction may be subject to local planning conditions such as improving site access, providing an equipment shed and lighting, etc. Planning authorities may also require a waiver to allow future use of a new tower by other operators at a reasonable rent. Soil sampling may be required depending on the design of tower footings - the structural engineer involved in tower design will need to advise on this requirement. Physical BS antenna deployment - Positioning antennas, waterproof cables and securing all connections with the aim of minimising future maintenance costs. Subscriber Deployment Subscriber agreement - A subscriber agreement will be required to define the terms and conditions of the service being provided. Besides the Implementing Wireless MANs obvious aspects such as cost and the duration of the contract, the agreement should cover any quality of service guarantees, liabilities, termination conditions, etc. An Internet search will turn up a wide variety of sample subscriber agreements, including many for wireless ISPs, although once again it is recommended to take legal advice to ensure that agreements comply with local laws. Subscriber site survey - Once the BS is installed, a simple site survey can be conducted at each new subscriber site by using the intended CPE antenna and receiver mounted on a light pole to lift the antenna to the intended installation point, with a cable length comparable to that required to run back to the subscribers computer location. A preliminary link budget calculation, followed by this type of quick site survey, can assure the quality of the wireless link and help prevent repeat visits in the event of performance problems. Customer premises equipment installation - If ease and speed of installation have been considered when selecting equipment, CPE installation can be prevented from consuming too much time and money as the network grows. Customer premises equipment grounding - Effective CPE grounding is important for three reasons, to make sure the antenna

is operating efficiently, to comply with any applicable local electrical installation regulations and for lightning protection. Connecting to the building's ground conduit will usually be sufficient, but local building regulations should be consulted to ensure compliance. Operating Phase Some of the key considerations that will require attention in the operating phase are described below. Technical Operations Customer helpline - The low cost start-up option is to use an answering service to help customers with basic instructions, FAQs and to point them to other more technically oriented help resources. If subscribers need support on more general networking/computing issues, an advanced technical support service could be provided but this may consume a significant slice of monthly revenues. Chapter Thirteen Subcontracting CPE and BS installation - Operator self-installation of CPE is the low-cost option for start-up, but a trained team of part-time installers, who are able to work flexible hours, will be an ideal solution once the venture is up and running. Scheduling and managing the team is a task that can be automated using e-mail and a suitable scheduling system. Business and Financial Operations Subscriber billing - Many off-the-shelf software systems are available for ISP billing. Typically these systems are designed to handle ISP specific features such as variable rates (flat, tiered, time/day or usage based), free offers (hours or usage), pre-paid card support and e-mail reports, reminders or invoices. Incentives - Can be an effective way to attract new subscribers. Existing subscribers can be given incentives to refer new customers, and others, such as installation contractors, can also be encouraged to promote the business by the incentive of extra work. Managing network operating costs - Controlling ongoing costs and achieving revenue targets will be a key focus in the operational phase. A clear understanding of the make-up of total network operations costs will be important, covering elements such as i

nstallation, infrastructure surveillance, maintenance, backhaul and administration costs. Business accounts - Options include DIY, with many software systems available, or hiring a part-time accountant who can either train on an existing accounting system or set one up for the venture. A small-scale WMAN venture should require no more than a day or two a month to manage all the business accounts. Summary of Part V Wireless metropolitan area networking is set for future growth following the publication of the initial IEEE 802.16 suite of standards and the progress towards mobility and mesh networking under development by Task Groups TGe, TGf and TGg. These standards provide the essential networking capabilities required in WMAN applications, scalability, service flexibility and quality of Implementing Wireless MANs service - capabilities that are beyond simpler MACs such as that specified in the original IEEE 802.11 standard. Standard compliance and interoperability is being progressed by the WiMAX Forum, and as certified products start to emerge, the promise of ubiquitous wireless broadband access that motivated the original IEEE 802.16 development will begin to be realised. This page intentionally left blank THE FUTURE OF WIRELESS NETWORKING TECHNOLOGY Introduction Besides showing the current status of wireless networking technologies, Parts III to IV have highlighted the areas where, over the full wireless range from a few centimetres to many kilometres, the further development and enhancement of these technologies is continuing. Aspects such as quality of service, roaming and satisfying the ever increasing demand for data bandwidth are the focus areas of many current developments. In this chapter some key developments are outlined that go beyond the incremental enhancement of existing wireless networking concepts. These developments, such as cognitive radio and media independent handoff, typically bridge across distance scales and are likely to be significant drivers of change in the fundamental nature of wire

less networking in the coming years. This page intentionally left blank CHAPTER Leading Edge Wireless Networking Technologies In this chapter four key technologies are discussed that are currently under development and that will play a large role in shaping the future of wireless networking; Wireless mesh network routing Network independent roaming Gigabit wireless LANs Cognitive, or spectrum agile radio. While individually these are significant step-outs from existing technologies (for example, cognitive radio radically extends the 802.11h enhancements to 802.11a networks that were described in the Section "Spectrum Management at 5 GHz (802.11h), p. 160"), together they herald a not-too-distant future in which spectrum availability, propagation range and data bandwidth are no longer limiting factors for wireless network performance. Wireless Mesh Network Routing As briefly described in the Section "Mesh Networks, p. 43", wireless mesh networks or mobile ad-hoc networks (MANETs) offer some significant advantages for large-scale wireless networking, including; self-organising architecture, optimising routing and traffic distribution Chapter Fourteen self-healing ability to respond to broken or unreliable wireless links increasing network throughput as the density of devices increases. A major challenge faced in defining mesh networking standards is to design Data Link layer protocols that are able to achieve this flexibility without consuming an excessive amount of network bandwidth for routing and control messaging. Simple, low overhead approaches to this problem, such as making routing decisions based on RF signal strength or the minimum hop count from source to destination, do not perform well compared to routing algorithms that actively probe the mesh topology and make routing decisions based on the historical and predicted throughput of the available paths through the mesh. One intriguing approach being investigated for mesh network routing is inspired by the communication method that enables ants to converge

on an optimum route to food sources, while maintaining back-up routes that can be used in the case of overcrowding or other obstacles. Ants use a method of communication called stigmergy, in which each ant modifies its local environment by laying down a chemical trail of pheromones and other ants respond to these modifications in such a way that the global behaviour of the colony becomes coordinated. Because the pheromone is volatile, and the intensity decays naturally with time, short, fast and often used routes will have a higher pheromone intensity than long, slow, blocked or abandoned routes, and will therefore be more often used and reinforced. Adaptation and improvement of existing routes, as well as the discovery of new routes, arises as a result of a degree of randomness inherent in the process. Some key elements of MANET routing algorithms inspired by this biological system are summarised in Table 14-1. While some ant colony inspired routing algorithms use either reactive or proactive strategies alone to gather information, combining these two approaches, and adding stochastic routing, results in a system that more closely mimics biological ant behaviour. It remains a challenge to minimise the bandwidth overhead used by the route sensing ant-agents, but algorithms that explicitly imitate aspects of ant behaviour may prove to be a key enabler for large-scale mesh networking. Leading Edge Wireless Networking Technologies Table 14-1: Features of Ant Colony Inspired MANET Routing Algorithms Routing feature Description Pheromone tables A table of routing information maintained in each node of the mesh, which indicates the "goodness" of the link to each of its neighbours in terms of data packet delivery time and number of hops to a destination. Reactive routing Software agents, unsurprisingly called ants, are generated information gathering to update pheromone tables in response to events such as a new station joining the mesh or the failure of a previously reliable route. Typically a "forward ant" seeks out

a route from source to destination and a "backward ant" returns over this route updating the tables in each intermediate node. Proactive routing Ants are periodically generated to proactively sample and information gathering optimise existing routes as well as discover alternative routes. Pheromone tables are updated to continuously optimise and evolve routing decisions as well as to respond to disruptive events. Stochastic routing When several alternative paths are available for a data decisions packet's next hop, a path is selected stochastically, giving higher probability to the path with the highest pheromone table value. This leads to automatic load balancing, since data is distributed across all good paths, and if one becomes overloaded it will be avoided until the congestion eases. Network Independent Roaming Media Independent Handoff In the Section "Network Performance and Roaming (802.11k and 802.11r), p. 162", three situations were described where client stations need to make transitions between WLAN access points; for mobile client stations moving out of range of a current access point, to maintain service availability under changing environmental conditions or service needs, or for load balancing within the WLAN. Transitions between points of attachment (POA) for a single network type (such as an 802.11 WLAN) are termed homogeneous transitions, and in the case of 802.11 networks, Task Groups TGk and TGr are developing and enhancing the mechanisms that enable seamless WLAN transitions. Chapter Fourteen Table 14-2: Roaming Needs Requiring Heterogeneous Transitions Roaming need Description Mobile client; A client station may move out of range of its current POA, and coverage need to transition to another network type because the current type is no longer available; e.g. moving out of range of an 802.11 hotspot and transitioning to a cellular phone service to maintain a voice connection. Mobile client; A mobile client may move into range of an alternative network cost advantage that is able to provide the

same or better QoS as the current POA but at a lower price; e.g. transitioning a voice call from a cellular phone service to a VoIP service when moving into range of an 802.11 hotspot. New service A new application is started that requires a level of service that requirement is not supported by the current POA; e.g. downloading a large file may be able to take advantage of a higher data rate available on another network. The next step in providing uninterrupted connectivity to the mobile user is to be able to make similar seamless transitions across multiple wireless networks of different types. These so-called heterogeneous transitions might involve a single user connection successively handing-off from an 802.11 WLAN to a cellular phone service and then to a WiMAX MAN and finally back to a WLAN via a new access point. As for homogeneous transitions, there are a number of reasons why mobile users may want to make this type of heterogeneous transition, as shown in Table 14-2. The challenge of enabling uninterrupted, QoS guaranteed, heterogeneous hand-offs is being addressed in the IEEE 802.21 Working Group, which started work in March 2004. 802.21 defines media independent handover (MIH) mechanisms that enable networks such as Wi-Fi, WiMAX and cellular phone networks to co-operate at the Network and Data Link layers of the OSI protocol stack (see Figure 14-1). The MIH function is a unified technology-independent interface that provides inputs and context to the upper layers to assist in handover decision-making. In turn the MIH gathers the necessary information on link parameters such as uplink /downlink rate, signal strength and range, and link capabilities such as QoS and security. This information is Leading Edge Wireless Networking Technologies Applications: Multimedia streaming, Internet, e-mail, voice MIH Messages : inter-radio handoff 802.11 WiMAX Cellular Network selection protocol protocol protocol Manager stack stack stack Network discovery 802.21 Smart triggers Wi-Fi WiMAX WWAN/Cellular Figure 14-1: M

IH Function in the Protocol Stack of a Multi-Radio Mobile Device gathered through technology-specific Layer 2 service access points for each of the enabled technologies, such as 802.11, 802.16 and 3GPP/3GPP2 for cellular phone networks. MIH in Practice The first devices and service that are putting MIH into practice are targeting the personal telephony sector, enabling cellular phone subscribers to use a single device to access cellular services when on the move and VoIP services when at home. BT's Fusion service, launched in the UK in 2005, uses Bluetooth enabled Motorola handsets together with a Bluetooth hub as a gateway for VoIP calls connected via a BT ADSL broadband Internet connection. Launched in partnership with Vodaphone, this service allows calls to be handed off between the cellular network and the VoIP over Bluetooth connection when the handset comes within range of the Bluetooth hub. Motorola announced a family of products in early 2006 that enable handoff between cellular phone and 802.11 based VoIP services. The residential seamless mobility gateway (RSG - Figure 14-2) includes an 802.11b/g access point, a four-port router and a VoIP adapter, allowing seamless handoff of voice calls between the home WLAN and the cellular network when using a dual-mode handset. These devices provide the expected VoIP features, such as 802.11i security and voice traffic prioritisation on the WLAN to ensure QoS, Chapter Fourteen Figure 14-2: Motorola Residential Seamless Mobility Gateway (Courtesy of Motorola Corporation) as well as offering many digital phone features, such as supporting multiple lines, caller ID, call waiting and call forwarding services. Ahead of the full development and ratification of the 802.21 standard, these devices rely on proprietary software and protocols to achieve a limited degree of media independent handoff and, in the BT case, the handoff only works with specific service providers. Nevertheless, these early demonstrations of the concept provide a foretaste of the remarkable flexibilit

y that full MIH will provide. Gigabit Wireless LANs The Section "MIMO and data rates to 600 Mbps (802.11n), p. 165" described how the 802.11 Task Group TGn is working towards delivering a PHY layer data rate of 500-600 Mbps, and an effective MAC SAP rate of 100 Mbps, through modifications to the 802.11 PHY and MAC layers and the application of MIMO radio. A number of standards based and proprietary equipment development projects are also underway aiming to deliver a PHY layer data rate of 1 Gbps, a technology threshold becoming known as Gi-Fi that is motivated by a range of home, office and public usage scenarios (Table 14-3). One such effort is the WIGWAM project, which stands for wireless gigabit with advanced multimedia, and was initiated in 2003 by a group of European companies and academic institutions to develop the enabling technologies for gigabit WLANs. The WIGWAM project is industry-driven rather than standards-driven, although the consortium intends to present its results to the relevant standards organisations. Leading Edge Wireless Networking Technologies Table 14-3: Gigabit Wireless Usage Scenarios Usage scenario Description Home usage Multiple concurrent high bandwidth media streaming applications (Video and HDTV) requiring data rates of several 100 Mbps per user. Fast synchronisation of personal storage devices exceeding 100 GB capacity. Office usage Replacing wired Ethernet in supporting high bandwidth office applications such as high-quality video conferencing, streaming media and network file sharing, with the required security and QoS. Public access Short range, very high data rate complement to existing public usage access networks such as GSM, GPRS, Wi-Fi, WiMAX, with seamless media independent handover. High speed Multi-user broadband Internet and media streaming to cars mobile usage and trains, with the additional technical challenge of varying Doppler shifts. The technical challenges faced in reaching this next step in wireless network throughput are familiar from earlier parts of the boo

k; maximising spectral efficiency to get more data bits into each hertz of RF spectrum maximising MAC efficiency SO that most of the transmitted bits are upper layer data rather than Link Control and MAC layer overhead ensuring security through strong encryption algorithms, with fast computation on low-cost hardware. In addressing these challenges, the WIGWAM consortium is considering broadly similar approaches to those under discussion by 802.11 TGn, for example, MIMO radio and OFDM with higher efficiency coding (LDPC and Turbo Codes - see next section) are under consideration to achieve high spectral efficiency. One additional technology being considered by WIGWAM is OFDMA with multi-carrier code division multiple access (OFDMA/MC-CDMA), described in the Section "Multi-Channel Code Division Multiple Access (MC-CDMA), p. 353". WIGWAM is initially targeting operation at 5 GHz but extensions in the 17, 24, 38 and 60 GHz RF bands are also under consideration. This would Chapter Fourteen bring wireless networking into the millimetre wave bands, with wavelengths of 8 mm at 38 GHz to 5 mm at 60 GHz. LDPC and Turbo Codes Low density parity check codes are error correction codes that are computationally efficient and perform close to the theoretical limit in enabling error recovery in noisy communication channels. Unlike a standard parity check, which simply flags bit errors during transmission, these codes also allow errors to be probabilistically recovered i.e. from a set of possible transmitted data blocks it is possible to compute which is most likely to have resulted in the received data block and check code. A check code is computed from a data block by sparsely sampling bits in the block using a random sampling matrix. Figure 14-3 illustrates a sampling matrix and the 4-bit check code resulting from the example 8-bit input work. The "low density" in LDPC refers to the fact that there are few 1's in the matrix. In this example, if the input word is incorrectly received as 00100101, the check code word can be used

to confirm that the most significant bit should have been a 1. X1X2XXXXXXXXXXXX 10100110 C1 =X 00011011 C2=X4X50 X8 11011000 C3=XX20X X5 01100101 C4=X2@X3@XX@XX Error correction step 2 1st and 3rd bits of the Sampling Matrix recomputed code word reversed by flipping the Transmitted code word incorrectly received bit X1 X1X2X3X4X5XXXXXX 10100101 C2=0@0@0@1 = 1 1000000 = Error correction step 1 Example input word C4=0@1@1@1 = 1 1st and 3rd bits of the recomputed code word differ from the received Recomputed code word XXXXXXXXXXXXXXXX code word C100101@0 = 0 00100101 C2=0 X = 1 C3=0@0@O@O = 0 Incorrect decoded word Figure 14-3: LDPC Computation and Error Correction Leading Edge Wireless Networking Technologies Turbo codes are another form of error correction code in which two parity checks are performed on the data block, one on the straight data and one on a known permutation of the data. In the receiver, two decoders use the two parity check blocks to compute the most likely sequence of transmitted bits. If the two decoders come up with different results, they exchange information and re-compute the most likely transmitted bit sequence, iterating until the two results are identical. The advantage of turbo codes is that they achieve very effective error recovery while keeping the coding rate close to 1, while the disadvantages are computational complexity and latency - in view of the iterative decoding process. Multi-Channel Code Division Multiple Access (MC-CDMA) As described in the Section "Code Division Multiple Access, p. 94", CDMA assigns an orthogonal Walsh-Hadamard code to each receiver and uses this as a chipping code to spread the input data stream. The orthogonality of chipping codes ensures that each receiver is only able to decode symbols encoded with that user's unique code. In MC-CDMA (Figure 14-4) each chip is transmitted in parallel using the same number of carriers as there are chips in the chipping code. In this case the orthogonality of codes allows multiple reuse of the same set of OFDM subcarri

ers by several concurrent users. Schematic block diagrams of an MC-CDMA transmitter and receiver are shown in Figure 14-5. From the left, a series to parallel converter splits the input bit stream into N parallel stream and each of these is further split into M parallel chip streams by the XOR operation with the M chips of the code word (C1 to CM). This results in the input bit stream being spread over a total of N X M subcarriers. The parallel chip streams of multiple users are XOR'd together and a modulator maps each resulting chip stream onto the amplitude/phase constellation in use, chip by chip for BPSK or in 6-chip symbols for 64-QAM. The N X M amplitude and phase points drive the inputs of an Inverse FFT and the computed output signal is transmitted after insertion of a guard interval. At the receiver, after guard interval removal the FFT computes the amplitude and phase of each of the N X M subcarriers and a demodulator Chapter Fourteen Chipping code, user A A, n-3 A, n-2 A, n-1 A, in A, n-3 B, n-3 Subcarrier n-3 C, n-3 Chipping code, user B D, n-3 B, n-3 B, n-2 B, n-1 B, n A, n-2 B, n-2 Chipping code, user C Subcarrier n-2 C, n-2 C, n-3 C, n-2 C, n-1 C, n D, n-2 A, n-1 Chipping code, user D B, n-1 Subcarrier n-1 1 C, n-1 D, n-3 D, n-2 D, n-1 D, n-1 in-chip Chipped user codes user data Chipped data spread over n sub-carriers Figure 14-4: Data Spreading in OFDMA/MC-CDMA Input Guard interval insertion stream Transmission channel M parallel chip streams from nth user Output Guard interval removal stream Figure 14-5: Schematic MC-CDMA Transmitter and Receiver Leading Edge Wireless Networking Technologies translates these constellation points into the equivalent input chips or multi-chip symbols. The M chips of the receiver's code word are XOR'd with each set of M demodulated chip streams to recover the N parallel bit streams which are finally converted back to a serial bit stream. Gigabit Wireless in Practice Siemens AG, a member of the WIGWAM consortium, announced the first 1 Gbps wireless link operating in

the 5 GHz band in December 2004, based on a 4 X 3 (Tx X Rx) MIMO radio together with an unspecified OFDM method. By June 2005, a data rate in excess of 10 Gbps was demonstrated by a University of Essex team in the UK, over a 60m line-of-sight link. Three RF bands between 2 and 7 GHz were used to create three concurrent data channels of 1.2, 1.6 and 2.4 Gbps, with each band also supporting a second concurrent channel using polarisation-based frequency reuse. Broad commercial application of 1 Gbps wireless LANs may be anticipated around 2010. Cognitive Radio The concept of a cognitive radio was first introduced by Joseph Miltola and Gerald Maguire in 2000, to represent a wireless device that combines an awareness of the RF environment together with learning and reasoning algorithms that can modify wireless PHY parameters in order to meet user requirements within the constraints of the RF environment. Spectrum agile radios are similar but with the emphasis on spectrum sensing and adaptation rather than on learning and reasoning. Spectrum sensing devices and algorithms are used to detect other users and enable spectrum agile radios to adjust their transmission parameters in response to the presence of other radios. Spectrum agile radios may also exchange sensing data in order to co-operate in making use of transmission opportunities. Another key concept underlying cognitive radio is that of a software defined radio (SDR) in which the digital signal processing functions such as data coding and modulation, are performed in software rather Chapter Fourteen than hardware. This gives a cognitive radio the flexibility to use alternative processing schemes, depending on changing requirements. Radio Frequency Policy Modernisation In December 2002, an FCC Spectrum Policy Review Task Force concluded with a number of recommendations aimed at modernising the regulatory framework to increase access to the RF spectrum, while assuring the freedom of existing licensed services from interference. This review was motivated by the cont

inually increasing demands on limited spectrum resources, coupled with the observation that, even in those parts of the RF spectrum that are fully allocated, the bandwidth is typically in use only 10-20% of the time. Following on from this review, the FCC issued a so-called notice of proposed rulemaking (NPRM) in May 2004, proposing to open up the 76-698 MHz portion of the TV spectrum in the US for unlicensed usage. This ruling would allow wireless networks to make unlicensed use of the unused "white space" in these licensed TV broadcast bands. Two types of devices are permitted to make unlicensed use of these bands under this ruling, as shown in Table 14-4. This opportunity led to the start up of the IEEE 802.22 Working Group in October 2004, which aims to develop MAC and PHY layer specifications for wireless regional area networks (WRAN) to operate in unused VHF/UHF TV bands between 54 and 862 MHz. Spectrum agile radio is the main enabling technology for the 802.22 work. Table 14-4: FCC Permitted Devices for Unlicensed Operation in TV Bands Device type Characteristics Fixed devices Maximum transmit power of 1W Must be either professionally installed to operate in locally unused channels or fitted with GPS and a means of determining which channels are locally vacant. This could be an Internet connection to a Spectrum Policy Server which provides location base spectrum coordination. Mobile devices Maximum transmit power of 100 mW Must receive a control signal from a device that determines which channels are vacant. Leading Edge Wireless Networking Technologies Spectrum Sharing Approaches and Challenges Approaches to spectrum sharing are broadly categorised as either vertical or horizontal. Vertical, also known as primary, sharing occurs when a channel is identified that is not being used - a so-called "white space" - while horizontal, also known as secondary, sharing is the efficient coexistence of two or more networks simultaneously operating in the same spectrum. Horizontal sharing relies either on rule based c

oordination of multiple radios within a single device, such as the Wi-Fi plus Bluetooth coexistence measures described in the Section "Bluetooth Coexistence with 802.11 WLANs, p. 247", or the coordination of radios via a separate common spectrum coordination channel (CSCC) that connects devices using beacon broadcasts in an edge-of-band channel, or by some other means, including potentially via the Internet. The 802.22 approach is based on vertical sharing, and the key challenge here is spectrum sensing - reliable dynamic detection of the presence of a primary (licensed) user operating in or returning to a channel. Three possible approaches to spectrum sensing are summarised in Table 14-5. Table 14-5: Spectrum Sensing Approaches for Vertical Spectrum Sharing Spectrum sensing Characteristics approach Matched receiver A receiver matched to the signal characteristics of the primary user (coding, modulation and synchronisation) and with knowledge of other signal characteristics and channel usage (spreading codes, pilots and training sequences, etc.). This provides very reliable detection but is relatively inflexible, although this can be partly overcome using a software defined radio. Energy detection Similar to a spectrum analyser, the receiver determines the presence or absence of a primary user by aggregating energy detected in the target channel. This can be effective for narrow band signals, but DSSS, FHSS or other wideband signals may not be detected. Feature detection Periodic features of modulated signals are detected by computing a spectral correlation, which detects periodicity from pulse trains, hopping cycles, cyclic prefixes, etc. Chapter Fourteen Using these sensing approaches, a spectrum agile radio would build up a local spectrum map, identifying the detected power level, signal type (e.g. FM, FHSS, OFDM) and bandwidth in use in each channel, as well as estimating the quality of empty channels in terms of their potential As an alternative to each spectrum agile radio making decisions based on its loca

l spectrum map, multiple radios could also collaborate in a vertical sharing mode, aggregating channel measurements via a CSCC protocol, in order to identify and exploit "white space" opportunities. If multiple radios compete for transmit opportunities, the CSCC channel would enable the contention to be resolved via a priority system, or alternatively dynamic auctions could be implemented, allowing devices to bid for media access. A further issue in spectrum sharing is identifying and preserving redundant "stand-by" channels, in case a primary user resumes broadcasting and the spectrum agile radio has to shift channels in order to avoid interfering. Solving these MAC and PHY layer challenges, starting with the efforts of the 802.22 Working Group in defining the specifications for wireless regional area networks, will enable cognitive and spectrum agile radios to open up the potential of a duty cycle approaching 100% in every hertz of the RF spectrum, unblocking spectrum availability as a constraint for a long time to come. Summary of Part VI The four developing technologies briefly described in this chapter promise a not-too-distant future in which wireless networks become truly ubiquitous - with virtually unlimited bandwidth available over a variety of different network types and ranges, enabling always on, mobile connectivity at any desired service level. The technology challenges of bandwidth, media access, QoS and mobility, have been recurring themes as each scale of wireless networking has been explored in earlier chapters. These new technologies may look set to consign these particular challenges to the history books, but there can be little doubt that new usage models Leading Edge Wireless Networking Technologies and services will also emerge that will put greater demands on wireless networks and pose new challenges for the standards, software and hardware developers. It is likely still to be some time before the final chapter can be written on wireless networking technology. This page intentionally left b

lank WIRELESS NETWORKING INFORMATION RESOURCES Introduction Part VII provides a knowledge base on wireless networking in three sections; a quick reference guide to some of the key online information sites and resources, arranged by wireless networking standard a comprehensive listing of acronyms commonly used in wired and wireless networking a glossary covering some of the key technical terms introduced earlier in the text. This page intentionally left blank CHAPTER Further Sources of Information The following sections aim to provide a quick reference to some of the key information sites and resources relating to wireless networking, where the latest up-to-date information on the status and further developments of the various standards can be followed. Any attempt to provide a comprehensive listing of available information on wireless networking in print form is destined to become quickly obsolete and overshadowed by the power of Google. Resources listed here have therefore been selected based on an expectation of longevity and continued relevance. General Information Sources Standards organisations Institute of Electrical and Electronic www.ieee.org/portal/site Engineers IEEE Wireless Standards Zone standards.ieee.org/wireless Technology fora Ultra Wideband Forum www.uwbforum.org The Wireless Association www.ctia.org Ultra Wideband Planet www.ultrawidebandplanet.org Wireless Communications Association www.wcai.com Resources RFC archive www.faqs.org/rfcs Chapter Fifteen NIST WLAN Security Framework www.src.nist.gov/pcig/checklists Wireless Net Design Line www.wirelessnetdesignline.com Wireless Design Online www.wirelessdesignonline.com Wireless Networking Tutorials www.wirelessnetworkstutorial.info Wireless Technology Information www.radio-electronics.co.uk/info/wireless Trade publications Wireless Week www.wirelessweek.com Wi-Fi Net News www.wifinetnews.com Wireless News Factor www.wirelessnewsfactor.com Mobile Enterprise www.mobilenterprisemag.com Wireless PAN Resources by Standard Bluetooth (IEEE 802.15.1) Sta

Resources Palowireless Bluetooth Resource Centre www.palowireless.com/bluetooth The Unofficial Bluetooth Weblog bluetooth.weblogsinc.com Blueserker - Berserk About Bluetooth www.blueserker.com Bluetooth Shareware www.bluetoothshareware.com News Tooth www.newstooth.com/newstooth Bluetooth tutorial www.tutorial-reports.com/wireless/bluetooth Suppliers Directory of Bluetooth products and services www.thewirelessdirectory.com/Bluetooth.htm Further Sources of Information Blueunplugged www.blueunplugged.com Ericsson www.ericsson.com/bluetooth Nokia www.nokia.com/bluetooth Motorola www.motorola.com/bluetooth Wireless USB Standards group WiMedia Alliance www.wimedia.org UWB Forum www.uwbforum.org Trade organisations USB Implementers' Forum www.usb.org Resources Palowireless UWB/Ultra Wideband www.palowireless.com/uwb Resource Centre USB-IF WUSB resources www.usb.org/developers/wusb Suppliers Staccato Communications www.staccatocommunications.com/products Belkin www.belkin.com Freescale www.freescale.com Wisair www.wisair.com ZigBee (IEEE 802.15.4) Standards group IEEE 802.15 Working Group www.ieee802.org/15 Trade organisations ZigBee Alliance www.zigbee.org Resources Palowireless ZigBee Resource Centre www.palowireless.com/zigbee Ultrawideband Insider www.uwbinsider.com ZigBee tutorial info www.tutorial-reports.com/wireless/zigbee Chapter Fifteen Suppliers Telegesis www.telegesis.com Crossbow Technology www.xbow.com Freescale www.freescale.com Cirronet www.cirronet.com Standards group Infrared Data Association www.irda.org Trade organisations www.irda.org Resources Palowireless IrDA/Infrared Resource www.palowireless.com/irda Centre www.eg3.org/irda.htm Suppliers ACTiSYS www.actisys.com Clarinet Systems www.clarinetsys.com FireWire (IEEE 1394) Standards group IEEE 1394 Working Group grouper.ieee.org/groups/1394/c IEEE 802.15.3 WPAN Working Group www.ieee802.org/15/pub/TG3.html Trade organisations IEEE 1394 Trade Association www.1394ta.org Resources Palowireless 802.15 WPAN Resource www.palowireless.com/i802_15 Centre App

le Computer Inc. developer.apple.com/devicedrivers/ firewire Further Sources of Information Suppliers FireWire Depot www.fwdepot.com/thestore Global Sources www.globalsources.com/manufacturers/ IEEE-1394-Firewire.html Near Field Communications (NFC) Standards group www.ecma-international.org Trade organisations NFC Forum www.nfc-forum.org Resources Radio Electronics NFC overview www.radio-electronics.com/info/wireless/ nfc/nfc_overview.php UNIK RFID tutorial wiki.unik.no/index.php/ Rfidtutorial Suppliers Philips www.semiconductors.philips.com/products/ identification/nfc Nokia www.nokia.com/nfc www.sony.net/Products/felica Wireless LAN Resources by Standard Wi-Fi (IEEE 802.11) Standards group IEEE 802.11 Working Group www.ieee802.org/11 Trade organisations Wi-Fi Alliance www.wi-fi.org Wireless LAN Association www.wlana.org Enhanced Wireless Consortium www.enhancedwirelessconsortium.org Chapter Fifteen Resources Palowireless 802.11 WLAN www.palowireless.com/i802_11 Resource Center Wi-Fi Planet www.wi-fiplanet.com 802.11 News www.80211anews.com Wireless Gumph www.wireless.gumph.org Wi-Fi tutorial info www.tutorial-reports.com/wireless/ wlanwifi Homemade LAN antennas www.wlan.org.uk/antenna-page.html Suppliers Proxim www.proxim.com Linksys www.linksys.com D-Link www.dlink.com Belkin www.belkin.com Netgear www.netgear.com PC based spectrum analyser www.cognio.com WLAN analyser www.netstumbler.com Bluetooth interference analyser www.airmagnet.com/products/ bluesweep.htm Site survey and WLAN planning tools www.wirelessvalley.com Wireless Mesh (IEEE 802.11s) Standards group IEEE 802.11 Working Group www.ieee802.org/11 Trade organisations Wi-Mesh www.wi-mesh.org Resources Mobile Pipeline tutorial www.mobilepipeline.com/howto/ 21600011 BelAir Networks resources www.belairnetworks.com/ resources Further Sources of Information Suppliers (proprietary, pre- 802.11s) BelAir Networks www.belairnetworks.com Nortel Networks www.nortelnetworks.com Tropos Networks www.tropos.com HiperLAN/2 (ETSI) Standards group European Telecommun

ications www.etsi.org Standards Institute Trade organisations HiperLAN2 Global Forum www.hiperlan2.com Resources Palowireless HiperLAN and www.palowireless.com/hiperlan2 HiperLAN/2 Resource Center Wireless MAN Resources by Standard WiMAX (IEEE 802.16) Standards group IEEE 802.16 Wireless MAN www.wirelessman.org Working Group Trade organisations WiMAX Forum www.wimaxforum.org WiMAX Industry www.wimax-industry.com Resources WiMax.com www.wimax.com 802.16 News www.80216news.com WiMaxxed www.wimaxxed.com Palowireless IEEE 802.16 WMAN www.palowireless.com/i802_16 Resource Center WiMAX tutorial info www.tutorial-reports.com/wireless/wimax Chapter Fifteen Starting, operating and maintaining www.startawisp.com WISPs. WISP Centric www.wispcentric.com Link budget tools www.wirelessconnections.net Suppliers Proxim www.proxim.com ACTiSYS www.actisys.com Solecktek Corporation www.solectek.com Antenna suppliers www.andrew.com Cognitive Radio Standards group FCC cognitive radio technologies www.fcc.gov/oet/cognitiveradio Trade organisations Software Defined Radio Forum www.sdrforum.org Resources Rutgers cognitive radio www.winlab.rutgers.edu/~xjing/prj/ resources CognitiveRadio.htm Programmable Wireless www.programmablewireless.org Suppliers Adapt4 www.adapt4.com GNU software radio www.gnu.org/software/gnuradio VANU software radio WWW.vanu.com CHAPTER Glossary A comprehensive guide to acronyms and other common terms in use in networking and wireless networking. Networking and Wireless Networking Acronyms Adaptive (or Advanced) Antenna System Access Category Acknowledge (Flow control frame) Asynchronous Connection-Less Advanced Encryption Standard Adaptive Frequency Hopping Arbitrary Inter Frame Spacing Access Point Adaptive Power Control Association of Radio Industries and Businesses Adaptive (or Automatic) Rate Selection Aggregate Server Access Protocol Amplitude shift keying Alternating Wireless Medium Access Chapter Sixteen Bit Error Rate Binary Phase Shift Keying Broadband Radio Access Networks Basic Rate Interface Base Sta

tion Basic Service Set BSSID Basic Service Set Identifier CBC-MAC Cyclic Block Chaining - Message Authentication Code Complementary Code Keying Counter Mode CBC-MAC Protocol Challenge-Handshake Authentication Protocol Connection or Channel Identifier Carrier to Interference Noise Ratio Cyclic Redundancy Check Common Spectrum Coordination Channel Channel State Information CSMA/CA Carrier Sensing Media Access/Collision Avoidance CSMA/CD Carrier Sensing Media Access/Collision Detection Clear to Send Decibels relative to an isotropic antenna Decibels relative to a 1 mW power level DBPSK Differential Binary Phase Shift Keying Distributed Coordination Function Dual Carrier Modulation Glossary Data Encryption Standard Dynamic Frequency Selection Dynamic Host Configuration Protocol DCF Inter Frame Spacing Data Link Control Differential Phase Shift Key DQPSK Differential Quadrature Phase Shift Keying Distributed Reservation Channel Access Distributed Reservation Protocol Dynamic Rate Shifting Distribution System Digital Subscriber Line Data Set Ready Direct Sequence Spread Spectrum Extensible Authentication Protocol EAPoL Extensible Authentication Protocol over LAN Electronic Code Book European Computer Manufacturers Association Enhanced Distributed Channel Access Enhanced Distributed Coordination Function Enhanced Data Rate (Bluetooth radio) Equivalent Isotropic Radiated Power Extended Service Set European Telecommunications Standards Institute Extended Unique Identifier Enhanced Wireless Consortium Chapter Sixteen Frame Check Sequence Frequency Division Duplex Forward Error Correction Fixed Frequency Interleaving Fast Fourier Transform Frequency Hopping Spread Spectrum Fast Infrared (IrDA) Frequency Shift Keying (also Fixed Symmetric Key) Fixed Wireless Access Gaussian Frequency Shift Keying Grant per Connection Grant per Subscriber Station HCF Controlled Channel Access Hybrid Coordination Function Host Controller Interface H-FDD Half duplex Frequency Division Duplexing HIPERLAN High Performance Radio Local Area Network

s HIPERMAN High Performance Radio Metropolitan Area Networks HIPERLAN 2 Inter-Integrated Circuit bus Internet Control Message Protocol Integrity Check Value Information Element Glossary Institute of Electrical and Electronic Engineers Internet Engineering Task Force Inter Frame Spacing Internet Protocol IPSec Internet Protocol Security Impulse Radio Infrared International Roaming Access Protocol IrCOMM Infrared COM port emulation Infrared Data Association IrDA Lite Reduced version of IrDA code IrLAN Infrared Local Area Network Protocol IrLAP Infrared Link Access Protocol IrLMP Infrared Link Management Protocol IrOBEX Infrared Object Exchange Protocol IrTran-P Infrared Image Exchange Protocol IrXfer Infrared File Transfer Protocol Integrated Services Digital Network Inter Symbol Interference IS-IS Intermediate System to Intermediate System International Standards Organisation International Telecommunications Union Initialisation Vector L2CAP Logical Link Control and Adaptation Protocol Layer 2 Tunnelling Protocol Local Area Network Chapter Sixteen Low Density Parity Check Logical Link Control Local Multipoint Distribution Service Link Manager Protocol Line of Sight Link Quality Indicator Least Significant Bit LWAPP Lightweight Access Point Protocol Media Access Control (also Message Authentication Code) MAC SAP MAC Service Access Point Metropolitan Area Network MANET Mobile Ad-hoc Network Media Access Slot MB-OFDM Multiband OFDM Multi-band OFDM Alliance Mesh Coordination Function Message Integrity Check Media Independent Handover Multiple Input, Multiple Output Micro-scheduled Management Command MAC Protocol Data Unit Most Significant Bit MS-CTA Micro-scheduled Channel Time Allocation MAC Service Data Unit Maximum Transmission Unit Glossary Network Address Translation Near Field Communications Non Line-of-Sight Network Operating System Non Return to Zero Inverted Offset Code Book Orthogonal Frequency Division Multiplexing OFDMA Orthogonal Frequency Division Multiple Access Open System Interconnect Open Shortest Pa

th First Personal Area Network Pulse Amplitude Modulation Password Authentication Protocol Port Address Translation Packet Binary Convolution Coding Prioritised Contention Access Point Coordination Function PCMCIA Personal Computer Memory Card International Association Protocol Data Unit Packet Error Rate Physical Layer PCF Interframe Spacing Public Key Infrastructure Point to Multipoint Chapter Sixteen Point of Attachment Personal Operating Space Point to Point Protocol Primary Rate Interface Pseudo Random Noise Pulse Shape Modulation Pair-wise Temporal (or Transient) Key Quadrature Amplitude Modulation Quality of Service Quadrature Phase Shift Keying RADIUS Remote Authentication Dial-In User Service Rivest (or Ron's) Code 4 Radio Frequency Request for Comments Radio Frequency Identification Routing Information Protocol Radio Link Control Rivest Shamir Adleman Robust Security Network Received Signal Strength Indicator Request to Send Return to Zero Inverted Synchronous Connection-Oriented Space Division Multiplexing Glossary Space Division Multiple Access Service Data Unit Simple, Efficient and Extensible Mesh Short Interframe Spacing Serial IrDA Server Message Block Simple Network Management Protocol Signal to Noise Ratio Small Office Home Office Simultaneously Operating Piconets Serial Peripheral Interface Spam over Internet Telephony Subscriber Station Service Set Identifier ST(B)C Space Time (Block) Coding Traffic Class TCP/IP Transport Control Protocol / Internet Protocol Time Division Duplex Time-Division Multiplexed Time Division Multiple Access Time Frequency Code Time Frequency Interleaving Task Group Tiny-P Flow control protocol for IrLMP connections Temporal Key Integrity Protocol Transport Layer Security Chapter Sixteen Tunnelled TLS Transmitter Power Control Transition Security Network Transmit Opportunity Universal Asynchronous Receiver Transmitter User Datagram Protocol U-NII Unlicensed - National Information Infrastructure Universal Serial Bus Unshielded Twisted Pair Ultra Wide Band Very Fast Inf

rared (IrDA) Virtual Local Area Network Voice over Internet Protocol VoWLAN Voice over Wireless Local Area Network Virtual Private Network Wireless Application Environment Wireless Application Protocol Wireless Distribution System Wired Equivalent Privacy WIGWAM Wireless Gigabit With Advanced Multimedia Support WiMAX Worldwide Interoperability for Microwave Access Wireless Internet Service Provider Wireless Local Area Network Glossary Wireless Metropolitan Area Network Wi-Fi Multimedia Wi-Fi Protected Access Wireless Personal Area Network Wireless Regional Area Network Wireless Robust Authenticated Protocol Wireless USB generic Digital Subscriber Line (ADSL, etc.) Networking and Wireless Networking Glossary Access Point: A wireless networking device that acts as a wireless hub and typically connects a wireless LAN to a wired network or to the Internet. Ad-hoc Mode: A wireless networking mode, also referred to as peer-to-peer mode or peer-to-peer networking, in which wireless enabled devices communicate with each other directly, without using an access point as a communication hub. See also Infrastructure Mode. Adaptive Burst Profiling: In adaptive burst profiling, transmission parameters such as modulation and coding schemes are adjusted on a frame by frame basis to meet the needs of individual client stations. This is a key element of the 802.16 MAC that provides flexibility for a wide range of services and devices in metropolitan area networking. Adaptive Frequency Hopping (AFH): Adaptive Frequency Hopping limits the channels used by a frequency hopping spread spectrum (FHSS) device to avoid channels being used by other co-located devices. AFH is applied in the 2.4 GHz ISM band to enable Bluetooth and Wi-Fi devices to co-operate without interference. Asymmetric Digital Subscriber Line (ADSL): ADSL is one of the number of technologies that enables higher bandwidth over standard copper Chapter Sixteen telephone lines. ADSL can achieve data rates of 640 kbps in the uplink and 9 Mbps in the downlink direction, with

a range of 6 km from the phone exchange. Asynchronous: In an asynchronous service, such as a generic file transfer, the transmission of packets within a data stream can be separated by random intervals. This compares with the more stringent time requirements of isochronous or synchronous services (q.v.). Backhaul: Backhaul refers to the transmission of network traffic from a remote site, such as a wireless MAN base station, back to an ISP or other Internet point-of-presence. A DSL link might provide backhaul for an urban Wi-Fi hotspot, while a remote rural network might require a long- range wireless or satellite link. Bandwidth: Bandwidth is a measure either of the frequency width of a signal (measured in Hertz), or the total amount of data that can be transmit- ted in a certain period of time over a particular medium or using a particu- lar device (measured in bps). The bandwidth of a transmitted signal is measured between the frequencies at which the signal drops to half its peak power (the 3dB points). Barker Code: A Barker code is a binary sequence that has low correlation with a time-shifted version of itself (low auto-correlation). The 11-bit Barker code (10110111000) is used as the spreading or chipping code in 802.11 DSSS. Baseband: Baseband refers to the signal in a communication system before it is modulated and multiplexed onto its carrier. The term is commonly used to refer both to the signal and to the hardware/software that processes the signal. Beamwidth: The coverage angle of a radio antenna, ranging from 360 degrees for an omnidirectional antenna to a narrow pencil beam with a high gain directional antenna such as a Yagi or parabolic dish. Binary Phase Shift Keying (BPSK): A modulation technique, using two phases of the carrier to represent data symbols 1 and 0. Bluetooth: A wireless PAN technology used for voice and data links with a typical range of 10 metres. Bluetooth delivers a standard data rate of 720 kbps Glossary using frequency hopping spread spectrum in the 2.4 GHz ISM radio band. Wi

th extended data rate (EDR) 2 Mbps and 3 Mbps rates can be achieved. Bonding: The process of link creation, pairing and authentication that occurs between Bluetooth devices. Bridge: A bridge is a link between two networks, for example a point-to- point wireless link connecting two wired networks. Broadcast: A broadcast message is transmitted to all receivers or stations connected to the network, in contrast to multicast or unicast messages (q.v.). Beacon messages used in many types of wireless networks are an example of broadcast messages. Carrier Sensing Media Access/Collision Avoidance: CSMA/CA is a method for multiple users to share access to the wireless medium while avoiding interfering with each other. A transmitter waiting to send data senses the medium to see if another station is transmitting, and uses a vari- ety of strategies, such as random back-off or RTS/CTS messages, to avoid collisions with other transmitters when the medium becomes free. Chipping Code: A code, such as a Barker or complementary code, used to spread a single bit into a longer sequence of chips to enable detection in a noisy communication channel. Coding Rate: The coding rate is an indication of the error-correction code overhead that is added to a data block to enable error recovery on reception. It is equal to m/(m+n) where n is the number of error correction bits applied to a data block of length m bits. Efficient error correction codes keep the coding rate close to unity. Complementary Code Keying (CCK): Complementary Code Keying is a type of Direct Sequence Spread Spectrum in which complementary codes (typically a set of 64 specific bit patterns) are used to encode the data stream and provide processing gain to enable detection of weak signals in a noisy environment. Connection Oriented/Connectionless Communication: Connection oriented communication takes place over a connection established between the Logical Link Control layers in the sending and receiving devices. Chapter Sixteen Link flow and error control mechanism are ava

ilable to ensure reliable and error free delivery for connection oriented communication. In contrast, con- nectionless communication proceeds without an LLC to LLC connection, when flow or error control are not required. Cyclic Redundancy Check: A cyclic redundancy check is the computa- tion for a block of data of a number, commonly called the checksum, which summarises and represents the content and organisation of bits in the input data block. By re-computing the CRC, the receiving station can detect any change, whether due to random transmission errors or to malicious inter- ception. In a CRC, the input data bits are taken as the coefficients of a poly- nomial. This polynomial is divided by another fixed polynomial and the CRC checksum bits are the coefficients of the remainder polynomial. The fixed divisor polynomial used in CRC-32 is x32 +x 26 + X 23 + 22 + X 16 + 8 7 + 2 +x+ 1, which can be represented in hexadecimal as either 04C11DB7 or EDB88320 depending on the bit-order convention (LSB first or MSB first). dBi: A logarithmic measure of the gain of an antenna relative to an isotropic antenna. dBm: A logarithmic measure of power relative to 1 milliwatt (mW). A power level of P dBm is ten to the power (P/10) milliwatts, SO that 20 dBm is 100 mW. Delay Spread: The delay spread is the variation in arrival time of wireless signals that propagate from a transmitter to a receiver over multiple paths. Inter symbol interference (ISI) will occur if consecutive symbols are transmitted closer together in time than the delay spread. Differential Binary Phase Shift Keying (DBPSK): A variation of the BPSK modulation technique in which a symbol is represented by a change of carrier phase rather than an absolute phase value. Differential Quadrature Phase Shift Keying (DQPSK): A variation of QPSK modulation in which a symbol is represented by a change of phase between two of the 4 QPSK constellation points, rather than the absolute phase value of one point. Digital Subscriber Line: A class of high speed Internet connectio

ns using standard telephone lines and delivering data rates of up to 1.5 Mbps. Glossary Directional Antenna: A directional antenna focuses transmitted power into a narrow beam, increasing the range at the expense of reduced angular coverage. Patch, Yagi and parabolic are types of directional antennas. Direct Sequence Spread Spectrum (DSSS): Direct Sequence Spread Spectrum is a data encoding technique in which the input bit stream is XOR'd with a chipping code to increase its bandwidth. On reception, the chipping code sequence is easier to detect in a noisy environment than a single transmitted bit, resulting in an additional gain in the system known as the processing gain. Longer chipping codes result in higher processing gain. Diversity: Diversity refers to a technique for improving the transmission of a signal, by receiving and processing multiple versions of the same trans- mitted signal. The multiple received versions can be the result of signals following different propagation paths (spatial diversity), being transmitted at different times (time diversity) or frequencies (frequency diversity). The simplest example of diversity in practice is the use of diversity anten- nas, where a receiver continuously senses the strength of received signals on two or more antennas and automatically selects the antenna receiving at maximum strength. Dynamic Host Configuration Protocol (DHCP): DHCP is a protocol that automatically provides network addressing and configuration information such as an IP address, sub-net mask and default gateway to a device when it connects to the network. Typically a device retains an assigned IP address for a specific administrator-defined period of time known as the lease period. Dynamic Frequency Selection (DFS): Dynamic Frequency Selection enables wireless networking devices to select a transmission channel to be used in order to avoid interference with other users, particularly radar and medical systems. DFS was introduced in the 802.11h supplement to enable 802.11a WLANs to comply with E

uropean regulations. Dynamic Routing: In contrast to static routing (q.v.), dynamic routing is the process whereby routers continuously update and exchange routing information in order to automatically adjust to changes in network topology. Equivalent Isotropic Radiated Power (EIRP): EIRP measures the total effective transmitted power of a radio, including antenna gain and any cable Chapter Sixteen and connector losses between transmitter and antenna. A 100 mW (20 dBm) transmitter with a 4 dB antenna and 1 dB of losses will have an EIRP of 200 mW (23 dBm). Ethernet: Ethernet is the predominant wired networking technology, developed at the Xerox Palo Alto Research Centre in the 1970s and standardised in the IEEE 802.3 specification. Ethernet types are designated a ABase-B, where A specifies the data rate, now up to 10 Gbps, and B specifies the cabling type, examples being T for twisted pair copper cabling and SX for LED powered multi-mode optical fibre. Extended Unique Identifier: An IEEE alternative to MAC addresses, extended in length from 48-bit to 64-bit addresses in EUI-64. Fading: Fading or multipath interference occurs when a primary signal combines with delayed signals, typically caused by reflection or refraction from objects on or near the line-of-sight, resulting in constructive or destructive (fading) interference or phase shifts. Fading can be identified and corrected by techniques such as the use of pilot tones in OFDM radios. Firewall: A firewall is a software component that controls the external interfaces of a network and restricts or blocks certain types of traffic or activity. Firewalls are essential for network security, and careful configura- tion is essential to ensure correct operation and to avoid unintended interruption of authorised network traffic. Forward Error Correction: Forward error correction (FEC) is a method of reducing bit errors during data transmissions. Redundant bits, each usu- ally a complex function of many bits from the input data stream, are added to the transmitted data

and enable the receiving device to detect and correct certain fraction of bit errors that occur during transmission. The fraction of original data bits in the final transmitted stream is called the coding rate. Frequency Reuse Factor: The frequency reuse factor is a measure of the degree to which a certain frequency can be used throughout the network, and therefore is a measure of the extent to which the full RF transmission bandwidth is available to carry data. For a large-scale 802.11b/g WLAN, with 3 non-overlapping channels patterned across adjacent access points, the frequency reuse factor is 1/3, as each channel is only used in 1/3 of the operating area of the network. Glossary Frequency Hopping Spread Spectrum (FHSS): Frequency Hopping Spread Spectrum is a spread spectrum technique, used in Bluetooth devices, in which the centre frequency of the modulated carrier hops peri- odically between a number of predetermined frequencies. The sequence of frequencies used is determined by a hopping code which is also known to the receiver. Fresnel Effect: A phenomenon in radio wave propagation in which an object that does not directly obstruct the line-of-sight from transmitter to receiver still causes attenuation of a transmission. The impact depends on how close the object is to the line-of-sight. Gateway: A gateway is a network component that connects one network to another, typically to the Internet. The gateway performs routing and pro- tocol translation (e.g. NAT, VPN passthrough), and may also provide addi- tional functions such as acting as a DHCP server. Hub: A hub is a network device that provides a central connection point for other devices. In contrast to a switch (q.v.), a hub broadcasts each data packet to every connected device, sharing the available bandwidth among all devices in the network. Infrastructure Mode: Infrastructure Mode is a mode of operation of a wireless network in which communication between devices takes place via an access point rather than directly between devices, as occurs in peer

-to- peer or ad-hoc mode (q.v.). Initialisation Vector (IV): An initialisation vector is part of the key used in an encryption algorithm, and is typically changed for each data packet. The IV is added to a secret key, such as a 40-bit key derived from a WEP passphrase, to obtain the full key used in the encryption algorithm. The IV is transmitted with the encrypted message and the receiving station, know- ing the secret key, can determine the full encryption key. The IV prevents the occurrence of patterns in the encrypted data that would otherwise make it easier for a hacker to determine the secret key. Chapter Sixteen Internet Protocol (IP): Internet Protocol is the Network layer protocol that provides addressing and routing functions on the Internet. The current version is IPv4. IP Address: A number that uniquely identifies a device on the Internet SO that other devices can communicate with it. IPv4 uses 32-bit addresses, while IPv6 addresses will be 128-bit. IP addresses are used by protocols at Layer 3 and above, and are translated into MAC addresses by the address resolution protocol (ARP) for use at Layer 2 and below. ISM (Industrial, Scientific and Medical): The ISM bands are three radio bands at 900 MHz, 2.4 and 5.8 GHz that were originally reserved for licence exempt use in industrial, scientific and medical applications, but are now also home to the main wireless networking PHY layer tech- nologies. Isochronous: An isochronous service requires data to be delivered within certain time constraints. Multimedia streams require an isochronous transport service to ensure that data frames are delivered as fast as they are displayed. The requirements of an isochronous service are not as rigid as those of a synchronous service, but are also not as lenient as an asynchro- nous service (q.v.). Jitter: Jitter refers to the variability in latency of individual transmitted packets, and is particularly important in determining quality of service for isochronous data services such as streaming video. Latency: The time

taken by a data packet to travel from its source to its destination. Latency is particularly important in some network services such as voice and video streaming, where transmission delays can seriously reduce the quality of service to the end user. Line-of-Sight (LOS): Line-of-sight refers to an unobstructed line between transmitting and receiving stations. Line-of-sight is required for any wire- less link for frequencies above ca. 11 GHz. When setting up long distance wireless links, a line-of-sight survey can be used to assess whether one aerial can be "seen" by another. Glossary Link Margin: The excess available power, over and above that required by the link budget, to achieve a desired received signal strength and SNR at the receiver. Local Area Network (LAN): A network used to link computers and other devices such as printers over short distances, typically tens of metres to a few hundred metres. The most common LAN technologies are 100 BaseT Ethernet for wired and 802.11b/g or Wi-Fi for wireless networks. Low Density Parity Check Codes: LDPC codes are error correction codes that are computationally efficient and perform close to the theoreti- cal limit in enabling error recovery in noisy communication channels. A check code is computed from a data block by sparsely sampling bits in the block using a random sampling matrix. For example, the first bit in the check code may be defined as y1 = X1 + X3 + 8 + 9 + 10, where the sum is modulus 2. The check code enables lost bits in the received data block to be recovered with high efficiency. MAC Service Access Point (MAC SAP): The MAC SAP is the logical point at the "top" of the MAC layer, where the MAC interfaces with the Network layer and higher layers in the OSI model. The MAC SAP is taken as a reference point when specifying the effective data rate of a network, as distinct from the data rate transmitted at the PHY layer. Manchester Coding: Manchester code is a data encoding method in which each bit or chip of data is represented by a transition between two