Datasets:

eswardivi

/

books

like 0

eld as an active device does. Instead a passive device transfers data back to an active device through a process called load modulation. These concepts are described in the following sections. Near Field and Far Field Communication The space around an antenna can be divided into two regions based on the differing nature of the electromagnetic fields generated by the antenna. The boundary between the two regions is known as the radian sphere and has a radius of 2/2, where n is the wavelength of the propagated electromagnetic wave. Radio Communication Basics The primary magnetic field begins at the antenna and oscillations in this field induce an electric field in the surrounding space. This region, inside the radian sphere, is within the influence of the primary magnetic field and is called the near field of the antenna. The electromagnetic field equations in this region reflect energy storage in the magnetic field and are described by near field coupling volume theory. The region outside the radian sphere is called the far field of the antenna and here the fields separate from the antenna and propagate into space as an electromagnetic wave. The electromagnetic field equations here represent energy propagation rather than storage, and propagation is described in terms of the concepts covered in the Section "RF Signal Propagation and Reception, p. 106". For NFC operating at 13.56 MHz, a = 22 metres, SO that the radius of the radian sphere is 2/2n = 3.5 metres. In the near field region, the magnetic field strength is inversely proportional to the cube of the distance between the antennas while the power in the magnetic field, which is used to energise passive NFC, decreases as the inverse sixth power of the separation. This is equivalent to an attenuation of 60 dB for a ten-fold increase in distance. Inductive Coupling Near-field inductive coupling uses an oscillating magnetic field to transfer RF energy between devices. Each device includes a resonant circuit tuned to the RF carrier frequency, and a loosely coupled

"space transformer" is established when the coil windings or "antenna loops" of the two devices are brought into range (Figure 4-38). The effective range is comparable to the actual physical dimensions of the transmitting antenna loop. When the resonant circuit in the transmitting device is energised by a RF power source, the resulting magnetic flux linkage results in energy transfer between the two resonant coil windings. Inductive coupling is only effective in the near-field region of the transmitting antenna loop. In the far-field region, where the electromagnetic field separates from the antenna and propagates as an electromagnetic wave, it can no longer have a direct effect through inductive coupling. Chapter Four Filter Magnetic field Oscillator Active NFC transmitter Passive NFC responder Figure 4-38: Inductive Coupling Between NFC Antenna Loops Load Modulation When an NFC target device in passive mode is within range of an active NFC transmitter (or initiator) its resonant circuit draws energy from the magnetic field created by the initiating device. This additional consumption results in a voltage perturbation that can be measured in the resonant circuit of the initiator. If an additional load resistance in the target is periodically switched on and off, this has the effect of an amplitude modulation of the carrier wave voltage in the initiator. By using the data stream to be transmitted from the target device to control this load switching, the data stream is transferred from the target to the initiator. This technique is called load modulation. Load modulation creates amplitude modulated sidebands on the 13.56 MHz carrier frequency, and the data stream is recovered by demodulating these sidebands in the initiating device's RF signal processing circuits. CHAPTER Infrared Communication Basics The Ir Spectrum The infrared (Ir) part of the electromagnetic spectrum covers radiation having a wavelength in the range from roughly 0.78 um to 1000 um (1 mm). Infrared radiation takes over from extremely high fre

quency (EHF) at 300 GHz and extends to just below the red end of the visible light spectrum at around 0.76 um wavelength. Unlike radio frequency radiation, which is transmitted from an antenna when excited by an oscillating electrical signal, infrared radiation is generated by the rotational and vibrational oscillations of molecules. The infrared spectrum is usually divided into three regions, near, middle and far, where "near" means nearest to visible light (Table 5-1). Although all infrared radiation is invisible to the human eye, far infrared is experienced as thermal, or heat, radiation. Rather than using frequency as an alternative to wavelength, as is commonly done in the RF region, the wavenumber is used instead in the infrared region. This is the reciprocal of the wavelength and is usually expressed as the number of wavelengths per centimetre. One aspect of wireless communication that becomes simpler outside the RF region is spectrum regulation, since the remit of the FCC and equivalent international agencies runs out at 300 GHz or 1 mm wavelength. Infrared Propagation and Reception The near infrared is the region used in data communications, largely as a result of the cheap availability of infrared emitting LEDs and optodetectors, Chapter Five Table 5-1: Subdivision of the Infrared Spectrum Infrared region Wavelength (um) Wavenumber (/cm) 0.78-2.5 12,800-4000 Middle 2.5-50 4000-200 50-1000 200-10 solid state devices that convert an electrical current directly into infrared radiation and vice versa. Infrared LEDs emit at discrete wavelengths in the range from 0.78 to 1.0 um, the specific wavelength of the LED being determined by the particular molecular oscillation that is used to generate the radiation. Transmitted Power Density - Radiant Intensity Ir propagation is generally a simpler topic than RF propagation, although the same principles, such as the concept of a link budget, still apply. As described in Chapter 4, the link budget predicts how much transmitter power is required to enable the received

data stream to be decoded at an acceptable BER. For Ir, the link budget calculation is far simpler than for RF, as terms like antenna gain, free space loss and multipath fading no longer apply. As a result, propagation behaviour can be more easily predicted for Ir than for RF. The unit of infrared power intensity, or radiant intensity, is mW/sr, with sr being the abbreviation for steradian. The steradian is the unit of solid angle measure, and this is the key concept in understanding the link budget for Ir communication. As shown in Figure 5-1, the solid angle (S) subtended by an area A on the surface of a sphere of radius R is given as: S=AIR2 steradians (5.1) A = 2 R2 (1-cos(a)) S = 2 (1-cos(a)) (5.2) Note from Eq. 5.1 that at a distance of 1 metre a solid angle of 1 steradian subtends an area of 1 m². For small solid angles, the area A on Infrared Communication Basics Radius of sphere = R Area = A Solid angle = A/R2 Figure 5-1: Solid Angle Subtended by an Area the sphere can be approximated by the area of the flat circle of radius r, giving: S =r2/R2steradians As an example, the IrDA physical layer standard specifies a half angle (a) of between 15° and 30°. For 15°, , S =2n (1-cos(15°)) = 0.214 steradians. For an LED with a given emitter power density or radiant intensity, I, in mW/sr, the equivalent power density in mW/m² will be approximately given as: P=1/R2mm/m2 (5.3) Emitter Beam Pattern Similar to an RF antenna, an LED has a beam pattern in which radiated power drops off with increasing angle off-axis. In the example shown in Figure 5-2, the power density drops to about 85% of the on-axis value at an off-axis angle of 15°. Inverse Square Loss Equation 5.3 shows that the on-axis power density is inversely proportional to the square of the distance from the source. If R is doubled, the power density P drops by a factor of 4, as shown in Figure 5-3. This is the equivalent of the free space loss term in the RF link budget. Chapter Five Offset angle Figure 5-2: Typical LED Emitted Power Polar Diagram Ir Detec

tor Sensitivity The standard detection device for high speed Ir communications is the photodiode, which has a detection sensitivity, or minimum threshold irradiance E, expressed in uW/cm2. In standard power mode (see the Section "IrDA PHY Layer, p. 282"), the IrDA standard specifies a minimum emitter power of 40 mW/sr. From Eq. 5.3, the minimum power density at a receiving photodiode at a range of 1 metre will be 40 mW/m², or 4 uW/cm². The sensitivity of a photo diode detector depends on the incident angle of the infrared source relative to the detector axis in a similar manner to that shown in Figure 5-2 for the beam pattern of an LED. Photodiode sensitivity also depends on the incident infrared wavelength as shown for example in Figure 5-4, and in any application a detector will be chosen Power density P = I & R 1 4 2 1g/2R2 Solid angle S Area = SR2 Source Distance from source Relative distance from source Figure 5-3: Inverse Square Distance Relationship of Radiant Power Density Infrared Communication Basics Wavelength (um) Figure 5-4: Typical Photodiode Sensitivity vs. Wavelength with a peak spectral sensitivity close to the wavelength of the emitting device. Ir Link Distance The maximum link distance for an Ir link can be calculated as the distance R at which the equivalent power density (P) drops to the level of the detector's minimum threshold irradiance (Ee). Eq. 5.3 gives: E = I / R2 mW/m² (5.4) The effective range of an Ir link can be increased substantially, up to several tens of metres, using lenses to collimate the transmitted beam and focus the beam onto the receiving photodiode. Alignment of the lenses and of the transmitting and receiving diodes will be critical to the effectiveness of such a system. As shown in Figure 5-5, a misalignment of approximately 1/3° would be sufficient to break a focussed link over a range of 10 m. Ir areal coverage may be increased in a home or small office environment by reflecting the Ir beam from a wall or ceiling in order to access a number of devices. In order to p

reserve power in the reflection it will be Chapter Five Angular offset 1/3° (exaggerated) Collimating Gathering source Beam offset 6cm detector at 10 metre distance Figure 5-5: Focussed In Link Alignment over a 10 m Range important to use a high reflectance, low absorption material to reflect the beam. A white painted ceiling is a good reflector of sunlight, with a reflection coefficient for visible light of around 0.9, but is a poor reflector at Ir wavelengths, absorbing about 90% of the incident Ir radiation. To achieve a comparable 0.9 reflection coefficient for Ir, an aluminium or aluminium foil covered panel would be a suitable reflector. Summary of Part II The radio frequency or infrared communication technologies described in Part II are at the heart of the physical layer of wireless networks. An understanding of the basics of these technologies will provide a firm foundation from which to discuss the implementation of wireless networks, whether local, personal or metropolitan area (LAN, PAN or MAN). In particular, the link budget calculation will be important in establishing power requirements and coverage in LAN and MAN applications. Spread spectrum and digital modulation techniques are key to understanding how a wide range of data rates can be accommodated within the limited available bandwidth, for example in the 2.4 and 5.8 GHz ISM bands. Ultra wideband radio - using a bandwidth approaching 7 GHz with a transmitted power density below the FCC allowable noise emission level - stretches the conceptual boundaries of what radio frequency communications can achieve, and the increasing number of practical applications including wireless USB and ZigBee, are testament to the practicality of this remarkable technology. Infrared Communication Basics Infrared communication links are perhaps the most "transparent" in terms of a very low requirement for user configuration, and to some extent this means that the user can be unconcerned about the underlying technology. However, even infrared links can be stretched t

o deliver performance over significant distances (tens of metres), given an understanding of the characteristics of Ir transmitters, detectors and infrared propagation. This page intentionally left blank WIRELESS LAN IMPLEMENTATION Introduction Wireless networking technology at the local area scale is perhaps the most widespread, most commercially significant and most well developed of all wireless networking technologies. In less than a decade since the ratification of the IEEE 802.11a and b standards in 1999, some 200 million 802.11 chipsets have been shipped, a business sector has grown to an estimated turnover in excess of $800 million (chipset cost alone) in 2005, and the standard has developed to the point where a 600- fold increase in data capacity, vehicular speed roaming and mesh networking are now on the horizon. Part III looks at the technologies and practical considerations that underpin successful wireless LANs. The main technical features of wireless LAN standards are described in Chapter 6. This is an area now firmly dominated by the IEEE 802.11 standards, starting with the original 802.11b based Wi-Fi, and now including the improved security features of 802.11i and the upcoming enhancements to throughput and mobility in progress with 802.11n, r and S. Non-IEEE WLAN standards are also reviewed, the brevity of this section being a clear indication of the dominance of the IEEE standards on the local area scale. Chapter 7 covers the implementation of WLANs, from the point of view of a medium scale corporate network, starting with the definition of user Part Three and technical requirements, through planning and installation to operation and support. This chapter concludes with a case study that looks at the specific requirements for a voice over WLAN (VoWLAN) application. Technologies providing security for wireless LANs are described in Chapter 8, including the most recent encryption and authentication mechanisms of 802.11i. Practical WLAN security measures are then described, including checklists co

vering management, technical and operational security measures. The last chapter in Part III addresses WLAN troubleshooting, and covers strategies for problem identification and diagnosis, as well as specific measures for the two most common categories of WLAN problems - connectivity and performance. CHAPTER Wireless LAN Standards The 802.11 WLAN Standards Origins and Evolution The development of wireless LAN standards by the IEEE began in the late 1980s, following the opening up of the three ISM radio bands for unlicensed use by the FCC in 1985, and reached a major milestone in 1997 with the approval and publication of the 802.11 standard. This standard, which initially specified modest data rates of 1 and 2 Mbps, has been enhanced over the years, the many revisions being denoted by the addition of a suffix letter to the original 802.11, as for example in 802.11a, b and g. The 802.11a and 802.11b extensions were ratified in July 1999, and 802.11b, offering data rates up to 11 Mbps, became the first standard with products to market under the Wi-Fi banner. The 802.11g specification was ratified in June 2003 and raised the PHY layer data rate to 54 Mbps, while offering a degree of interoperability with 802.11b equipment with which it shares the 2.4 GHz ISM band. Table 6-1 summarises the 802.11 standard's relentless march through the alphabet, with various revisions addressing issues such as security, local regulatory compliance and mesh networking, as well as other enhancements that will lift the PHY layer data rate to 600 Mbps. Overview of the Main Characteristics of 802.11 WLANs The 802.11 standards cover the PHY and MAC layer definition for local area wireless networking. As shown in Figure 6-1, the upper part of the Data Link layer (OSI Layer 2) is provided by Logical Link Control (LLC) Chapter Six Table 6-1: The IEEE 802.11 Standard Suite Standard Key features 802.11a High speed WLAN standard, supporting 54 Mbps data rate using OFDM modulation in the 5 GHz ISM band. 802.11b The original Wi-Fi standard, providi

ng 11 Mbps using DSSS and CCK on the 2.4 GHz ISM band. 802.11d Enables MAC level configuration of allowed frequencies, power levels and signal bandwidth to comply with local RF regulations, thereby facilitating international roaming. 802.11e Addresses quality of service (QoS) requirements for all 802.11 radio interfaces, providing TDMA to prioritise and error-correction to enhance performance of delay sensitive applications. 802.11f Defines recommended practices and an Inter-Access Point Protocol to enable access points to exchange the information required to support distribution system services. Ensures inter-operability of access points from multiple vendors, for example to support roaming. 802.11g Enhances data rate to 54 Mbps using OFDM modulation on the 2.4 GHz ISM band. Interoperable in the same network with 802.11b equipment. 802.11h Spectrum management in the 5 GHz band, using dynamic frequency selection (DFS) and transmit power control (TPC) to meet European requirements to minimise interference with military radar and satellite communications. 802.11i Addresses the security weaknesses in user authentication and encryption protocols. The standard employs advanced encryption standard (AES) and 802.1x authentication. 802.11j Japanese regulatory extension to 802.11a adding RF channels between 4.9 and 5.0 GHz. 802.11k Specifies network performance optimisation through channel selection, roaming and TPC. Overall network throughput is maximised by efficiently loading all access points in a network, including those with weaker signal strength. 802.11n Provides higher data rates of 150, 350 and up to 600 Mbps using MIMO radio technology, wider RF channels and protocol stack improvements, while maintaining backward compatibility with 802.11 a, b and g. 802.11p Wireless access for the vehicular environment (WAVE), providing commu- nication between vehicles or from a vehicle to a roadside access point using the licensed intelligent transportation systems (ITS) band at 5.9 GHz. Wireless LAN Standards Table 6-1: The

IEEE 802.11 Standard Suite - cont'd 802.11r Enables fast BSS to BSS (Basic Service Set) transitions for mobile devices, to support delay sensitive services such as VoIP on stations roaming between access points. 802.11s Extending 802.11 MAC to support ESS (Extended Service Set) mesh networking. The 802.11s protocol will enable message delivery over self-configuring multi-hop mesh topologies. 802.11T Recommended practices on measurement methods, performance metrics and test procedures to assess the performance of 802.11 equipment and networks. The capital T denotes a recommended practice rather than a technical standard. 802.11u Amendments to both PHY and MAC layers to provide a generic and standardised approach to inter-working with non-802.11 networks, such as Bluetooth, ZigBee and WiMAX. 802.11v Enhancements to increase throughput, reduce interference and improve reliability through network management. 802.11w Increased network security by extending 802.11 protection to management as well as data frames. services specified in the 802.2 standard, which are also used by Ethernet (802.3) networks, and provide the link to the Network layer and higher layer protocols. 802.11 networks are composed of three basic components; stations, access points and a distribution system, as described in Table 6-2. In the 802.11 standard, WLANs are based on a cellular structure where each cell, under the control of an access point, is known as a basic service Network layer Logical Link Control (LLC) Data Link layer Medium Access Control (MAC) Physical layer Physical layer (PHY) OSI Model layers 802.11 specifications Figure 6-1: 802.11 Logical Architecture Chapter Six set (BSS). When a number of stations are working in a BSS it means that they all transmit and receive on the same RF channel, use a common BSSID, use the same set of data rates and are all synchronised to a common timer. These BSS parameters are included in "beacon frames" that are broadcast at regular intervals either by individual stations or by the access point. The

standard defines two modes of operation for a BSS; ad-hoc mode and infrastructure mode. An ad-hoc network is formed when a group of two or more 802.11 stations communicate directly with each other with no access point or connection to a wired network. This operating mode (also known as peer-to-peer mode) allows wireless connections to be quickly established for data sharing among a group of wireless enabled computers (Figure 6-2). Under ad-hoc mode the service set is called an independent basic service set (IBSS), and in an IBSS all stations broadcast beacon packets, and use a randomly generated BSSID. Infrastructure mode exists when stations are communicating with an access point rather than directly with each other. A home WLAN with an access point and several wired devices connected through an Ethernet hub or switch is a simple example of a BSS in infrastructure mode (Figure 6-3). All communication between stations in a BSS goes through the access point, even if two wireless stations in the same cell need to communicate with each other. This doubling-up of communication within a cell (first from sending station to the access point, then from the access point to the Table 6-2: 802.11 Network Components Component Description Station Any device that implements the 802.11 MAC and PHY layer protocols. Access point A station that provides an addressable interface between a set of stations, known as a basic service set (BSS), and the distribution system. Distribution system A network component, commonly a wired Ethernet, that connects access points and their associated BSSs to form an extended service set (ESS). Wireless LAN Standards Independent Basic Service Set (IBSS) Figure 6-2: Ad-hoc Mode Topology destination station) might seem like an unnecessary overhead for a simple network, but among the benefits of using a BSS rather than an IBSS is that the access point can buffer data if the receiving station is in standby mode, temporarily out of range or switched off. In infrastructure mode, the access point takes on

the role of broadcasting beacon frames. The access point will also be connected to a distribution system which will usually be a wired network, but could also be a wireless bridge to other WLAN cells. In this case the cell supported by each access point is a BSS and if two or more such cells exist on a LAN the combined set is known as an extended service set (ESS). Distribution system connection Basic Service Set (BSS) Figure 6-3: Infrastructure Mode Topology Chapter Six In an ESS, access points (APs) will use the distribution system to transfer data from one BSS to another, and also to enable stations to move from one AP to another without any interruption in service. The transport and routing protocols that operate on the external network have no concept of mobility - of the route to a device changing rapidly - and within the 802.11 architecture the ESS provides this mobility to stations while keeping it invisible to the outside network. Prior to 802.11k, support for mobility within 802.11 networks was limited to movement of a station between BSSs within a single ESS, so-called BSS transitions. With 802.11k, which will be described further in the Section "Network Performance and Roaming (802.11k and 802.11r, p. 162)", the roaming of stations between ESSs is supported. When a station is sensed as moving out of range, an access point is able to deliver a site report that identifies alternative access points the station can connect to for uninterrupted service. The 802.11 MAC Layer The MAC layer is implemented in every 802.11 station, and enables the station to establish a network or join a pre-existing network and to transmit data passed down by Logical Link Control (LLC). These functions are delivered using two classes of services, station services and distribution system services, which are implemented by the transmission of a variety of management, control and data frames between MAC layers in communicating stations. Before these MAC services can be invoked, the MAC first needs to gain access to the wireless

medium within a BSS, with potentially many other stations also competing for access to the medium. The mechanisms to efficiently share access within a BSS are described in the next section. Wireless Media Access Sharing media access among many transmitting stations in a wireless network is more complex to achieve than in a wired network. This is because a wireless network station is not able to detect a collision between its transmission and the transmission from another station, since Wireless LAN Standards a radio transceiver is unable both to transmit and to listen for other stations transmitting at the same time. In a wired network a network interface is able to detect collisions by sensing the carrier, for example the Ethernet cable, during transmission and ceasing transmission if a collision is detected. This results in a medium access mechanism known as carrier sense multiple access/ collision detection (CSMA/CD). The 802.11 standard defines a number of MAC layer coordination functions to co-ordinate media access among multiple stations. Media access can either be contention-based, as in the mandatory 802.11 distributed coordination function (DCF), when all stations essentially compete for access to the media, or contention free, as in the optional 802.11 point coordination function (PCF), when stations can be allocated specific periods during which they will have sole use of the media. The media access method used by the distributed coordination function is carrier sense multiple access/collision avoidance (CSMA/CA), illustrated in Figure 6-4. In this mode a station that is waiting to transmit will sense the medium on the channel being used and wait until the medium is free of other transmissions. Once the medium is free, the station waits a predetermined period (the distributed inter-frame spacing or DIFS). If the station senses no other transmission before the end of the DIFS period, it computes a random backoff time, between parameter values CWmin and CWmax, and commences its transmission if the medium

remains free after this time has elapsed. The contention window parameter Cw is Device A Data packet Medium busy Medium free Medium free Device B attempts to send Carrier Random sensing Data packet backoff Medium busy Medium free Medium busy Device C attempts to send Carrier Random Device C computes a longer backoff time than device B sensing backoff Backoff counter is suspended when medium goes busy Figure 6-4: 802.11 CSMA/CA Chapter Six specified in terms of a multiple of a slot time that is 20 us for 802.11b or 9 us for 802.11a/g networks. The backoff time is randomised SO that, if many stations are waiting, they will not all try again at the same time - one will have a shorter backoff and will succeed in starting its transmission. If a station has to make repeated attempts to transmit a packet, the computed backoff period is doubled with each new attempt, up to a maximum value Cw max defined for each station. This ensures that, when many stations are competing for access, individual attempts are spaced out more widely to minimise repeated collisions. If another station is sensed transmitting before the end of the DIFS period, this is because a short IFS (SIFS) can be used by a station that is waiting either to transmit certain control frames (CTS or ACK - see Figure 6-5) or to continue the transmission of parts of a data packet that has been fragmented to improve transmission reliability. Device A Random backoff Destination A Device B Random backoff Random backoff Start of contention period Start of contention period Figure 6-5: DCF Transmission Timing CSMA/CA is a simple media access protocol that works efficiently if there is no interference and if the data being transmitted across the network is not time critical. In the presence of interference, network throughput can be dramatically reduced as stations continually backoff to avoid collisions or wait for the medium to become idle. CSMA/CA is a contention-based protocol, since all stations have to compete for access. With the exception of the SIFS mechani

association, it can use active scanning by transmitting a Probe frame containing this SSID and waiting for a Probe Response frame to be returned by the preferred access point. A broadcast Probe frame can also be sent, requesting all access points within range to respond with a Probe Response. This will provide the new station with a full list of access points available. The process of authentication and association can then start - either with the preferred access point or with another access point selected by the new station or by the user from the response list. Station Services MAC layer station services provide functions to send and receive data units passed down by the LLC and to implement authentication and security between stations, as described in Table 6-3. Chapter Six Table 6-3: 802.11 MAC Layer Station Services Service Description Authentication This service enables a receiving station to authenticate another station prior to association. An access point can be configured for either open system or shared key authentication. Open system authentication offers minimal security and does not validate the identity of other stations - any station that attempts to authenticate will receive authentication. Shared key authentication requires both stations to have received a secret key (e.g. a passphrase) via another secure channel such as direct user input. Deauthentication Prior to disassociation, a station will deauthenticate from the station that it intends to stop communication with. Both deauthentication and authentication are achieved by the exchange of management frames between the MAC layers of the two communicating stations. Privacy This service enables data frames and shared key authentication frames to be optionally encrypted before transmission, for example using wired equivalent privacy (WEP) or Wi-Fi protected access (WPA). MAC service data A MAC service data unit (MSDU) is a unit of data passed to the unit delivery MAC layer by the logical link controller. The point at which the LLC accesses MAC s

ervices (at the "top" of the MAC layer) is termed the MAC service access point or SAP. This service ensures the delivery of MSDUs between these service access points. Control frames such as RTS, CTS and ACK may be used to control the flow of frames between stations, for example in 802.11b/g mixed-mode operation. Distribution System Services The functionality provided by MAC distribution system services is distinct from station services in that these services extend across the distribution system rather than just between sending and receiving stations at either end of the air interface. The 802.11 distribution system services are described in Table 6-4. 802.11 PHY Layer The initial 802.11 standard, as ratified in 1997, supported three alternative PHY layers; frequency hopping and direct sequence spread Wireless LAN Standards Table 6-4: 802.11 MAC Layer Distribution System Services Service Description Association This service enables a logical connection to be made between a station and an access point. An access point cannot receive or deliver any data until a station has associated, since association provides the distribution system with the information necessary for delivery of data. Disassociation A station disassociates before leaving a network, for example when a wireless link is disabled, the network interface controller is manually disconnected or its host PC is powered down. Reassociation The reassociation service allows a station to change the attributes (such as supported data rates) of an existing association or to change its association from one BSS to another within an extended BSS. For example, a roaming station may change its association when senses another access point transmitting a stronger beacon frame. Distribution The distribution service is used by a station to send frames to another station within the same BSS, or across the distribution system to a station in another BSS. Integration Integration is an extension of distribution when the access point is a portal to a non-802.11 network and th

e MSDU has to be transmitted across this network to its destination. The integration service provides the necessary address and media specific translation SO that an 802.11 MSDU can be transmitted across the new medium and successfully received by the destination device's non-802.11 MAC. spectrum in the 2.4 GHz band as well as an infrared PHY. All three PHYs delivered data rates of 1 and 2 Mbps. The infrared PHY specified a wavelength in the 800-900 nm range and used a diffuse mode of propagation rather than direct alignment of infrared transceivers, as is the case in IrDA for example (Section 10.5). A connection between stations would be made via passive ceiling reflection of the infrared beam, giving a range of 10-20 metres, depending on the height of the ceiling. Pulse position modulation was specified, 16-PPM and 4-PPM respectively for the 1 and 2 Mbps data rates. Later extensions to the standard have focused on high rate DSSS (802.11b), OFDM (802.11a and g) and OFDM plus MIMO (802.11n). These PHY layers will be described in the following sections. Chapter Six 802.11a PHY Layer The 802.11a amendment to the original 802.11 standard was ratified in 1999 and the first 802. 1a compliant chipsets were introduced by Atheros in 2001. The 802.11a standard specifies a PHY layer based on orthogonal frequency division multiplexing (OFDM) in the 5 GHz frequency range. In the US, 802.11a OFDM uses the three unlicensed national information infrastructure bands (U-NII), with each band accommodating four non-overlapping channels, each of 20-MHz bandwidth. Maximum transmit power levels are specified by the FCC for each of these bands and, in view of the higher permitted power level, the four upper band channels are reserved for outdoor applications. In Europe, in addition to the 8 channels between 5.150 and 5.350 GHz, 11 channels are available between 5.470 and 5.725 GHz (channels 100, 104, 108, 112, 116, 120, 124, 128, 132, 136, 140). European regulations on maximum power level and indoor versus outdoor use vary from country

to country, but typically the 5.15-5.35 GHz band is reserved for indoor use with a maximum EIRP of 200 mW, while the 5.47-5.725 GHz band has an EIRP limit of 1W and is reserved for outdoor use. Table 6-5: US FCC Specified U-NII Channels Used in the 802.11 a OFDM PHY RF Band Frequency Channel Centre frequency Maximum Range number (GHz) transmit power (GHz) U-NII lower 5.150-5.250 5.180 5.200 5.550 5.240 U-NII middle 5.250-5.350 5.260 5.280 5.300 5.320 U-NII upper 5.725-5.825 5.745 5.765 5.785 5.805 Wireless LAN Standards As part of the global spectrum harmonisation drive following the 2003 ITU World Radio Communication Conference, the 5.470-5.725 GHz spectrum has also been available in the US since November 2003, subject to the implementation of the 802.11h spectrum management mechanisms described in the Section "Spectrum Management at 5 GHz (802.11h), p. 160". Each of the 20 MHz wide channels accommodates 52 OFDM subcarriers, with a separation of 312.5 kHz (= 20 MHz/64) between centre frequencies. Four of the subcarriers are used as pilot tones, providing a reference to compensate for phase and frequency shifts, while the remaining 48 are used to carry data. Four different modulation methods are specified, as shown in Table 6-6, which result in a range of PHY layer data rates from 6 Mbps up to 54 Mbps. The coding rate indicates the error-correction overhead that is added to the input data stream and is equal to m/(m+n) where n is the number of error correction bits applied to a data block of length m bits. For example, with a coding rate of 3/4 every 8 transmitted bits includes 6 bits of user data and 2 error correction bits. The user data rate resulting from a given combination of modulation method and coding rate can be determined as follows, taking the Table 6-6: 802.11a OFDM Modulation Methods, Coding and Data Rate Modulation Code bits per Code bits per Coding rate Data bits per Data rate subcarrier (Mbps) symbol symbol 16-QAM 16-QAM 64-QAM 64-QAM Chapter Six 64-QAM, 3/4 coding rate line as an example. Durin

g one symbol period of 4 uS, which includes a guard interval of 800 nS between symbols, each carrier is encoded with a phase and amplitude represented by one point on the 64-QAM constellation. Since there are 64 such points, this encodes 6 code bits. The 48 subcarriers together therefore carry 6 x 48 = 288 code bits for each symbol period. With a 3/4 coding rate, 216 of those code bits will be user data while the remaining 72 will be error correction bits. Transmitting 216 data bits every 4 uS corresponds to a data rate of 216 data bits per OFDM symbol X 250 OFDM symbols per second = 54 Mbps. The 802.11a specifies 6, 12 and 24 Mbps data rates as mandatory, corresponding to 1/2 coding rate for BPSK, QPSK and 16-QAM modulation methods. The 802.11a MAC protocol allows stations to negotiate modulation parameters in order to achieve the maximum robust data rate. Transmitting at 5 GHz gives 802.11a the advantage of less interference compared to 802.11b, operating in the more crowded 2.4 GHz ISM band, but the higher carrier frequency is not without disadvantages. It restricts 802.11a to near line-of-sight applications and, taken together with the lower penetration at 5 GHz, means that indoors more WLAN access points are likely to be required to cover a given operating area. 802.11b PHY Layer The original 802.11 DSSS PHY used the 11-chip Barker spreading code (as described in the Section "Chipping, Spreading and Correcting, p. 80") together with DBPSK and DQPSK modulation methods to deliver PHY layer data rates of 1 and 2 Mbps respectively (Table 6-7). The high rate DSSS PHY specified in 802.11b added complementary code keying (CCK), using 8-chip spreading codes, as described in the Section "Complementary Code Keying, p. 82". The 802.11 standard supports dynamic rate shifting (DRS) or adaptive rate selection (ARS), allowing the data rate to be dynamically adjusted to compensate for interference or varying path losses. When interference is present, or if a station moves beyond the optimal range for reliable operation at t

he maximum data rate, access points will progressively fall Wireless LAN Standards Table 6-7: 802.11b DSSS Modulation Methods, Coding and Data Rate Modulation Code length Code type Symbol rate Data bits Data rate (Chips) (Msps) per symbol (Mbps) Barker Barker DQPSK 1.375 DQPSK 1.375 back to lower rates until reliable communication is restored. This strategy is based on the implications of Eq. (4-1), which showed that SNR is proportional to the transmitted energy per bit, SO that by falling back to a lower data rate, a higher SNR and lower BER can be achieved. Conversely, if a station moves back within range for a higher rate, or if interference is reduced, the link will shift to a higher rate. Rate shifting is implemented in the PHY layer and is transparent to the upper layers of the protocol stack. The 802.11 standard specifies the division of the 2.4 GHz ISM band into a number of overlapping 22 MHz channels, as shown in Figure 4-9. The FCC in the US and the ETSI in Europe have both authorised the use of spectrum from 2.400 to 2.4835 GHz, with 11 channels approved in the US and 13 in (most of) Europe. In Japan, channel 14 at 2.484 GHz is also authorised by the ARIB. Some countries in Europe have more restrictive channel allocations, notably France where only four channels (10 through 13) are approved. The available channels for 802.11b operation are summarised in Table 6-8. The 802.11b standard also includes a second, optional modulation and coding method, packet binary convolutional coding (PBCCTT-Texas Instruments), which offers improved performance at 5.5 and 11 Mbps by achieving an additional 3 dB processing gain. Rather than the 2 or 4 phase states or phase shifts used by BPSK/DQSK, PBCC uses 8-PSK (8 phase states) giving a higher chip per symbol rate. This can be translated into either a higher data rate for a given chipping code length, or a higher processing gain for a given data rate, by using a longer chipping code. Chapter Six Table 6-8: International Channel Availability for 802.11b Networks in the 2

.4 GHz Band Channel number Centre frequency (GHz) Geographical usage 2.412 US, Canada, Europe, Japan 2.417 US, Canada, Europe, Japan 2.422 US, Canada, Europe, Japan 2.427 US, Canada, Europe, Japan 2.432 US, Canada, Europe, Japan 2.437 US, Canada, Europe, Japan 2.442 US, Canada, Europe, Japan 2.447 US, Canada, Europe, Japan 2.452 US, Canada, Europe, Japan 2.457 US, Canada, Europe, Japan, France 2.462 US, Canada, Europe, Japan, France 2.467 Europe, Japan, France 2.472 Europe, Japan, France 2.484 Japan 802.11g PHY Layer The 802.11g PHY layer was the third 802.11 standard to be approved by the IEEE standards board and was ratified in June 2003. Like 802.11b, 11g operates in the 2.4 GHz band, but increases the PHY layer data rate to 54 Mbps, as for 802.11a. The 802.11g uses OFDM to add data rates from 12 Mbps to 54 Mbps, but is fully backward compatible with 802.11b, SO that hardware supporting both standards can operate in the same 2.4 GHz WLAN. The OFDM modulation and coding scheme is identical to that applied in the 802.11a standard, with each 20 MHz channel in the 2.4 GHz band (as shown in Table 6-8) divided into 52 subcarriers, with 4 pilot tones and 48 data tones. Data rates from 6 to 54 Mbps are achieved using the same modulation methods and coding rates shown for 802.11a in Table 6-6. Wireless LAN Standards Although 802.11b and 11g hardware can operate in the same WLAN, throughput is reduced when 802.11b stations are associated with an 11g network (so-called mixed-mode operation) because of a number of protection mechanisms to ensure interoperability, as described Table 6-9. Table 6-9: 802.11b/g Mixed-Mode Interoperability Mechanisms Mechanism Description RTS/CTS Before transmitting, 11b stations request access to the medium by sending a request to send (RTS) message to the access point. Transmission can commence on receipt of the clear to send (CTS) response. This avoids collisions between 11b and 11g transmissions, but the additional RTS/CTS signalling adds a significant overhead that decreases network throu

ghput. CTS to self The CTS to self option dispenses with the exchange of RTS/CTS messages and just relies on the 802.11b station to check that the channel is clear before transmitting. Although this does not provide the same degree of collision avoidance, it can increase throughput significantly when there are fewer stations competing for medium access. Backoff time 802.11g backoff timing is based on the 802.11a specification (up to a maximum of 15x9 uS slots) but in mixed-mode an 802.11g network will adopt 802.11b backoff parameters (maximum 31 X 20 uS slots). The longer 802.11b backoff results in reduced network throughput. The impact of mixed mode operation on the throughput of an 802.11g network is shown in Table 6-10. A number of hardware manufacturers have introduced proprietary extensions to the 802.11g specification to boost the data rate above 54 Mbps. An example is D-Link's proprietary "108G" which uses packet bursting and channel bonding to achieve a PHY layer data rate of 108 Mbps. Packet bursting, also known as frame bursting, bundles short data packets into fewer but larger packet to reduce the impact of gaps between transmitted packets. Packet bursting as a data rate enhancement strategy runs counter to packet fragmentation as a strategy for improving transmission robustness, SO packet bursting will only be effective when interference or high levels of contention between stations are absent. Chapter Six Channel bonding is a method where multiple network interfaces in a single machine are used together to transmit a single data stream. In the 108G example, two non-overlapping channels in the 2.4 GHz ISM band are used simultaneously to transmit data frames. Data Rates at the PHY and MAC Layer In considering the technical requirements for a WLAN implementation in Chapter 7, it will be important to recognise the difference between the headline data rate of a wireless networking standard and the true effective data rate as seen by the higher OSI layers when passing data packets down to the MAC layer. Ea

ch "raw" data packet passed to the MAC service access point (MAC SAP) will acquire a MAC header and a message integrity code and additional security related header information before being passed to the PHY layer for transmission. The headline data rate, for example 54 Mbps for 802.11a or 11g networks, measures the transmission rate of this extended data stream at the PHY layer. The effective data rate is the rate at which the underlying user data is being transmitted if all the transmitted bits relating to headers, integrity checking and other overheads are ignored. For example, on average, every 6 bits of raw data passed to the MAC SAP of an 802.11b WLAN will gain an extra 5 bits of overhead before transmission, reducing a PHY layer peak data rate of 11 Mbps to an effective rate of 6 Mbps. Table 6-10 shows the PHY and MAC SAP data rates for 802.11 WLANs. For 802.11g networks the MAC SAP data rate depends on the presence of 802.11b stations, as a result of the mixed-mode media access control mechanisms described in the previous section. 802.11 Enhancements In the following sections some of the key enhancements to 802.11 network capabilities and performance will be described. The security enhancements covered by the 802.11i update are described separately in Chapter 8, which is devoted to WLAN security. Wireless LAN Standards Table 6-10: PHY and MAC SAP Throughput Comparison for 802.11a, b and g Networks Network standard and configuration PHY data Effective Effective rate (Mbps) MAC SAP Throughput throughput versus (Mbps) 802.11b (%) 802.11b network 802.11g network with 802.11b stations (CTS/RTS) 802.11g network with 802.11b stations (CTS-to-self) 802.11g network with no 802.11b stations 802.11a Quality of Service (802.11e specification) The 802.11e specification provides a number of enhancements to the 802.11 MAC to improve the quality of service for time sensitive applications, such as streaming media and voice over wireless IP (VoWIP), and was approved for publication by the IEEE Standards Board in September 2

005. The 802.11e specification defines two new coordination functions for controlling and prioritising media access, which enhance the original 802.11 DCF and PCF mechanisms described above in Section "Wireless Media Access, p. 144". Up to eight traffic classes (TC) or access categories (AC) are defined, each of which can have specific QoS requirements and receive specific priority for media access. The simplest of the 802.11e coordination functions is enhanced DCF (EDCF) which allows several MAC parameters determining ease of media access to be specified per traffic class. An arbitrary interframe space (AIFS) is defined which is equal to DIFS for the highest priority traffic class and longer for other classes. This provides a deterministic mechanism for traffic prioritisation as shown in Figure 6-6. The minimum backoff time CWmin is also TC dependent, SO that, when a collision occurs, higher priority traffic, with a lower CWmin, will have a higher probability of accessing the medium. Chapter Six Device A Interframe spacing for control frames and continuation of high priority transmission Device B Interframe spacing for Point Coordination Function Device C High priority traffic; DIFS and short contention period Device D AIFS[2] Medium priority traffic; AIFS[2] and intermediate contention period Device E AIFS[3] Low priority traffic; AIFS[3] and longer contention period Figure 6-6: EDCF Timing Each station maintains a separate queue for each TC (Figure 6-7), and these behave as virtual stations, each with their individual MAC parameters. If two queues within a station reach the end of their backoff periods at the same time, data from the higher priority queue will be transmitted when the station gains access to the medium. Although the EDCF coordination mode does not provide a guaranteed service for any TC, it has the advantage of being simple to configure and implement as an extension of DCF. The second enhancement defines a new hybrid coordination function (HCF) which complements the polling concept of PCF with

an awareness of the Application data Application frames assigned to traffic class queues Voice data Video data Best efforts Background Internal collision resolution Priority frame sent to PHY layer PHY layer Figure 6-7: EDCF Traffic Class Queues Wireless LAN Standards DIFS=34 us Voice SIFS 2 slots Backoff 0 3 slots Video DIFS=34 us SIFS 2 slots Backoff 0 7 slots Best effort AIFS[2]=43 us SIFS 3 slots Backoff 0 15 slots AIFS[3]=79 us Background Backoff 0 15 slots SIFS 7 slots Interframe spacing per traffic class Backoff period traffic class Figure 6-8: AIFS and Backoff Timing per WMM Traffic Class QoS requirements of each station. Stations report the lengths of their queues for each traffic class and the hybrid coordinator uses this to determine which station will receive a transmit opportunity (TXOP) during a contention free transmission period. This HCF controlled channel access (HCCA) mechanism considers several factors in determining this allocation; The priority of the TC The QoS requirements of the TC (bandwidth, latency and jitter) Queue lengths per TC and per station The duration of the TXOP available to be allocated The past QoS given to the TC and station. HCCA allows applications to schedule access according to their needs, and therefore enables QoS guarantees to be made. Scheduled access requires a client station to know its resource requirements in advance and scheduling concurrent traffic from multiple stations also requires the access point to make certain assumptions regarding data packet sizes, data transmission rates and the need to reserve surplus bandwidth for transmission retries. The Wi-Fi Alliance adopted a subset of the 802.11e standard in advance of the IEEE's September 2005 approval. This subset, called Wi-Fi multimedia (WMM), describes four access categories as shown in Table 6-11, with EDCF timings as shown in Figure 6-8. The prioritisation mechanism certified in WMM is equivalent to the EDCF coordination mode defined in 802.11e but did not initially include the scheduled access capabil

ity available through HCF and HCCA. This and other 802.11e capabilities are planned to be progressively included in the Wi-Fi Alliance's WMM certification program. Chapter Six Table 6-11: WMM Access Category Descriptions Access category Description WMM voice priority Highest priority. Allows multiple concurrent VoWLAN calls, with low latency and quality equal to a toll voice call. WMM video priority Prioritises video traffic above lower categories. One 802.11g or 802. 11a channel can support 3 to 4 standard definition TV streams or 1 high definition TV stream. WMM best effort priority Traffic from legacy devices, from applications or devices that lack QoS capabilities, or traffic such as internet surfing that is less sensitive to latency but is affected by long delays. WMM background priority Low priority traffic, such as a file download or print job, that does not have strict latency or throughput requirements. Spectrum Management at 5 GHz (802.11h) The 802.11h standard supplements the 802.11 MAC with two additional spectrum management services, transmit power control (TPC), which limits the transmitted power to the lowest level needed to ensure adequate signal strength at the farthest station, and dynamic frequency selection (DFS), which enables a station to switch to a new channel if it detects other non-associated stations or systems transmitting on the same channel. These mechanisms are required for 5 GHz WLANs operating under European regulations, in order to minimise interference with satellite communications (TPC) and military radar (DFS), and support for the 802.11h extensions was required from 2005 for all 802.1 11a compliant systems operating in Europe. In the US, compliance with 802.11h is also required for 802.11a products operating in the 12 channels from 5.47 to 5.725 GHz. IEEE 802.11h compliant networks therefore have access to 24 non-overlapping OFDM channels, resulting in a potential doubling of overall network capacity. Transmit Power Control An 802.11h compliant station indicates its transmit

power capability, including minimum and maximum transmit power levels in dBm, in the Wireless LAN Standards association or reassociation frame sent to an access point. An access point may refuse the association request if the station's transmit power capability is unacceptable, for example if it violates local constraints. The access point in return indicates local maximum transmit power constraints in its beacon and probe request frames. An access point monitors signal strength within its BSS by requesting stations to report back the link margin for the frame containing the report request and the transmit power used to transmit the report frame back to the access point. This data is used by the access point to estimate the path loss to other stations and to dynamically adjust transmit power levels in its BSS in order to reduce interference with other devices while maintaining sufficient link margin for robust communication. Dynamic Frequency Selection When a station uses a Probe frame to identify access points in range, an access point will specify in the Probe Response frame that it uses DFS. When a station associates or re-associates with an access point that uses DFS, the station provides a list of supported channels that enables the access point to determine the best channel when a shift is required. As for TPC, an access point may reject an association request if a station's list of supported channels is considered unacceptable, for example if it is too limited. To determine if other radio transmissions are present, either on the channel in use or on a potential new channel, an access point sends a measurement request to one or more stations identifying the channel where activity is to be measured, and the start time and duration of the measurement period. To enable these measurements, the access point can specify a quiet period in its beacon frames to ensure that all other associated stations stop transmission during the measurement period. After performing the requested measurement, stations send back a r

eport on the measured channel activity to the access point. When necessary, channel switching is initiated by the access point, which sends a channel switch announcement management frame to all associated stations. This announcement identifies the new channel, specifies the number of beacon periods until the channel switch takes effect, and also specifies whether or not further transmissions are allowed on the current channel. The access point can use the short interframe Chapter Six spacing (SIFS - see the Section "Wireless Media Access, p. 144") to gain priority access to the wireless medium in order to broadcast a channel switch announcement. Dynamic frequency selection is more complicated in an IBSS (ad-hoc mode) as there is no association process during which supported channel information can be exchanged, and no access point to coordinate channel measurement or switching. A separate DFS owner service is defined in 802.11h to address these complications, although channel switching remains inherently less robust in an IBSS than in an infrastructure mode BSS. Network Performance and Roaming (802.11k and 802.11r) A client station may need to make a transition between WLAN access points for one of three reasons, as described in Table 6-12. The 802.11 Task Groups TGk and TGr are addressing issues relating to handoffs or transitions between access points that need to be fast and reliable for applications such as VoWLAN. TGk will standardise radio measurements and reports that will enable location-based services, such as a roaming station's choice of a new access point to connect to, while TGr aims to minimise the delay and maintain QoS during these transitions. 802.11k; Radio Resource Measurement Enhancements The 802.11 Task Group TGk, subtitled Radio Resource Measurement Enhancements, began meeting in early 2003 with the objective of Table 6-12: Reasons for Roaming in a WLAN Roaming need Description Mobile client station A mobile client station may move out of range of its current access point and need to transit

ion to another access point with a higher signal strength. Service availability The QoS available at the current access point may either deteriorate or may be inadequate for a new service requirement, for example if a VoWLAN application is started. Load balancing An access point may redirect some associated clients to another available access point in order to maximize the use of available capacity within the network. Wireless LAN Standards defining radio and network information gathering and reporting mechanisms to aid the management and maintenance of wireless LANs. The 11k supplement will be compatible with the 802.11 MAC as well as implementing all mandatory parts of the 802.11 standards and specifications, and targets improved network management for all 802.11 networks. The key measurements and reports defined by the supplement are as follows; Beacon reports Channel reports Hidden station reports Client station statistics Site reports. IEEE 802.11k will also extend the 802.11h TPC to cover other regulatory requirements and frequency bands. Stations will be able to use these reports and statistics to make intelligent roaming decisions, for example eliminating a candidate access point if a high level of non-802.11 energy is detected in the channel being used. The 802.11k supplement only addresses the measurement and reporting of this information and does not address the processes and decisions that will make use of the measurements. The three roaming scenarios described above will be enabled by the TGk measurements and reports, summarised in Table 6-13. For example, a mobile station experiencing a reduced RSSI will request a neighbour report from its current access point that will provide information on other access points in its vicinity. A smart roaming algorithm in the mobile station will then analyse channel conditions and the loading of candidate access points and select a new access point that is best able to provide the required QoS. Once a new access point has been selected, the station will perform a

BSS transition by disassociating from the current access point and associating with the new one, including authentication and establishing the required QoS. Chapter Six Table 6-13: 802.11k Measurements and Reports 802.11k feature Description Beacon report Access points will use a beacon request to ask a station to report all the access point beacons it detects on a specified channel. Details such as supported services, encryption types and received signal strength will be gathered. Channel reports (noise Access points can request stations to construct a noise histogram, medium histogram showing all non-802.11 energy detected on a sensing time histogram specified channel, or to report data about channel loading report and channel (how long a channel was busy during a specified time load report) interval as well as the histogram of channel busy and idle times). Hidden station report Under 802.11k, stations will maintain lists of hidden stations (stations that they can detect but are not detected by their access point). Access points can request a station to report this list and can use the information as input to roaming decisions. Station statistic report 802.11k access points will be able to query stations to and frame report report statistics such as the link quality and network performance experienced by a station, the counts of packets transmitted and received, and transmission retries. Site report A station can request an access point to provide a site report - a ranked list of alternative access points based on an analysis of all the data and measurements available via the above reports. 802.11r; Fast BSS Transitions The speed and security of transitions between access points will be further enhanced by the 802.11 specification which is also under development and is intended to improve WLAN support for mobile telephony via VoWLAN. IEEE 802.11r will give access points and stations the ability to make fast BSS to BSS transitions through a four-step process; Active or passive scanning for other access points in

the vicinity, Authentication with one or more target access points, Reassociation to establish a connection with the target access point, and Pairwise temporal key (PTK) derivation and 802. 1x based authentication via a 4-way handshake, Wireless LAN Standards leading to re-establishment of the connection with continuous QoS through the transition. A key element of the process of associating with the new access point will be a pre-allocation of media reservations that will assure continuity of service - a station will not be in the position of having jumped to a new access point only to find it is unable to get the slot time required to maintain a time critical service. The 802.11k and 802.11r enhancements address roaming within 802.11 networks, and are a step towards transparent roaming between different wireless networks such as 802.11, 3G and WiMAX. The IEEE 802.21 media independent handover (MIH) function, which is described further in the Section "Network Independent Roaming, p. 347", will eventually enable mobile stations to roam across these diverse wireless networks. MIMO and Data Rates to 600 Mbps (802.11n) The IEEE 802.11 Task Group TGn started work during the second half of 2003 to respond to the demand for further increase in WLAN performance, and aims to deliver a minimum effective data rate of 100 Mbps through modifications to the 802.11 PHY and MAC layers. This target data rate, at the MAC service entry point (MAC SAP), will require a PHY layer data rate in excess of 200 Mbps, representing a fourfold increase in throughput compared to 802.11a and 11g networks. Backward compatibility with 11a/b/g networks will ensure a smooth transition from legacy systems, without imposing excessive performance penalties on the high rate capable parts of a network. Although there is still considerable debate among the supporters of alternative proposals, the main industry group working to accelerate the development of the 802.11n standard is the enhanced wireless consortium (EWC) which published Rev 1 of its MAC an

d PHY proposals in September 2005. The following description is based on the EWC proposals. The two key technologies that are expected to be required to deliver the aspired 802.11n data rate are multi-input multi-output (MIMO) radio and OFDM with extended channel bandwidths. Chapter Six MIMO radio, outlined in the Section "MIMO Radio, p. 124" is able to resolve information transmitted over several signal paths using multiple spatially separated transmitter and receiver antennas. The use of multiple antennas provides an additional gain (the diversity gain) that increases the receiver's ability to decode the data stream. The extension of channel bandwidths, most likely by the combination of two 20 MHz channels in either the 2.4 GHz or 5 GHz bands, will further increase capacity since the number of available OFDM data tones can be doubled. To achieve a 100 Mbps effective data rate at the MAC SAP it is expected to require either a 2 transmitter X 2 receiver antenna system operating over a 40 MHz bandwidth or a 4 X 4 antenna system operating over 20 MHz, with respectively 2 or 4 spatially separated data streams being processed. In view of the significant increase in hardware and signal processing complexity in going from 2 to 4 data streams, the 40 MHz bandwidth solution is likely to be preferred where permitted by local spectrum regulations. To ensure backward compatibility, a PHY operating mode will be specified in which 802.11a/g OFDM is used in either the upper or lower 20 MHz of a 40 MHz channel. Maximising data throughput in 802.11n networks will require intelligent mechanisms to continuously adapt parameters such as channel bandwidth and selection, antenna configuration, modulation scheme and coding rate, to varying wireless channel conditions. A total of 32 modulation and coding schemes are initially specified, in four groups of eight, depending on whether one to four spatial streams are used. Table 6-14 shows the modulation and coding schemes for the highest rate case - four spatial streams operating over 40

MHz bandwidth providing 108 OFDM data tones. For fewer spatial streams, the data rates are simply proportional to the number of streams. As for 802.11a/g, these data rates are achieved with a symbol period of 4.0 uS. A further data rate increase of 10/9 (e.g. from 540 to 600 Mbps) is achieved in an optional short guard interval mode, which reduces the symbol period to 3.6 uS by halving the inter-symbol guard interval from 800 nS to 400 nS. Wireless LAN Standards Table 6-14: 802.11n OFDM Modulation Methods, Coding and Data Rate Modulation Code bits per Code bits per Coding Data bits Data rate subcarrier symbol per symbol (Mbps) (per stream) (all streams) (all streams) 16-QAM 16-QAM 64-QAM 64-QAM 64-QAM MAC framing and acknowledgement overheads will also need to be reduced in order to increase MAC efficiency (defined as the effective data rate at the MAC SAP as a fraction of the PHY layer data rate). With the current MAC overhead, a PHY layer data rate approaching 500 Mbps would be required to deliver the target 100 Mbps data rate at the MAC SAP. Mesh Networking (802.11s) As described in the Section "The 802.11 WLAN Standards, p. 139", the 802.11 topology relies on a distribution system (DS) to link BSSs together to form an ESS. The DS is commonly a wired Ethernet linking access points (Figure 6-3), but the 802.11 standard also provides for a wireless distribution system between separated Ethernet segments by defining a four-address frame format that contains source and destination station addresses as well as the addresses of the two access points that these stations are connected to, as shown in Figure 6-9. The objective of the 802.11s Task Group, which began working in 2004, is to extend the 802.11 MAC as the basis of a protocol to establish a wireless distribution system (WDS) that will operate over self-configuring multi-hop wireless topologies, in other words an ESS mesh. Chapter Six LAN segment A 802.3 Ethernet Sta 4 Sta 1 frame Four address format 802.11 frame Sta 1 Sta 2 LAN segment B WDS link Sta 3 Sta 4

802.3 Ethernet frame Figure 6-9: Wireless Distribution System Based on Four-Address Format MAC Frame An ESS mesh is a collection of access points, connected by a WDS, that automatically learns about the changing topology and dynamically re-configures routing paths as stations and access points join, leave and move within the mesh. From the point of view of an individual station and its relationship with a BSS and ESS, an ESS mesh is functionally equivalent to a wired ESS. Two industry alliances emerged during 2005 to promote alternative technical proposals for consideration by TGs; the Wi-Mesh Alliance and SEEMesh (for Simple, Efficient and Extensible Mesh). The main elements of the Wi-Mesh proposal are a mesh coordination function (MCF) and a distributed reservation channel access protocol (DRCA) to operate alongside the HCCA and EDCA protocols (Figure 6-10). Some of the key features of the proposed Wi-Mesh MCF are summarised in Table 6-15. The final ESS mesh specification is likely to include prioritised traffic handling based on 802.11e QoS mechanisms as well as security features and enhancements to the 802.11i standard. The evolution of 802.11 security will be fully described in Chapter 8, but mesh networking introduces some security considerations in addition to those that have been progressively solved for non-mesh WLANs by Wireless LAN Standards Mesh Internetworking Measurement Routing Security Mesh Coordination Function (MCF) HCCA / EDCA 802.11n MAC Distributed Coordination Function (DCF) 802.11a/11b/11g/11j PHY 802.11n PHY 802.11s functions 802.11n functions Figure 6-10: Wi-Mesh Logical Architecture WEP, WPA, WPA2 and 802.11i. In a mesh network additional security methods are needed to identify nodes that are authorised to perform routing functions, in order to ensure a secure link for routing information messages. This will be more complicated to achieve in a mesh, where there will commonly be no centralised authentication server. The work of the 802.11s TG is at an early stage, and ratification of the

final accepted proposal is not expected before 2008. Table 6-15: Wi-MESH Mesh Coordination Function (MCF) Features Wi-Mesh MCF feature Description Media access coordination Media access coordination in a multi-hop network to across multiple nodes avoid performance degradation and meet QoS guarantees. Support for QoS Traffic prioritisation within the mesh; flow control over multi-hop paths; load control and contention resolution mechanisms. Efficient RF frequency To mitigate performance loss resulting from hidden and spatial reuse and exposed stations, and allow for concurrent transmissions to enhance capacity. Scalability Enabling different network sizes, topologies and usage models. PHY independent Independent of the number of radios, channel quality, propagation environment and antenna arrangement (including smart antennas). Chapter Six Other WLAN Standards Although the wireless LAN landscape is now comprehensively dominated by the 802.11 family of standards, there was a brief period in the evolution of WLAN standards when that dominance was far from assured. From 1998 to 2000, equipment based on alternative standards briefly held sway. This short reign was brought to an end by the rapid market penetration of 802.11b products, with 10 million 802.11b based chipsets being shipped between 1999 and end-2001. The HomeRF and HiperLAN standards, which are now of mainly historical interest, are briefly described in the following sections. HomeRF The Home Radio Frequency (HomeRF) Working Group was formed in 1998 by a group of PC, consumer electronics and software companies, including Compaq, HP, IBM, Intel, Microsoft and Motorola, with the aim of developing a wireless network for the home networking market. The Working Group developed the specification for SWAP - Shared Wireless Access Protocol - which provided wireless voice and data networking. SWAP was derived from the IEEE's 802.11 and ETSI's DECT (digitally enhanced cordless telephony) standards and includes MAC and PHY layer specifications with the main character

istics summarised in Table 6-16. Table 6-16: Main Characteristics of the HomeRF SWAP Main characteristics specification TDMA for synchronous data traffic - up to 6 TDD voice conversations. CSMA/CA for asynchronous data traffic, with prioritisation for streaming data. CSMA/CA and TDMA periods in a single SWAP frame. FHSS radio in the 2.4 GHz ISM band. 50-100 hops per second. 2- and 4-FSK modulation deliver PHY layer data rates of 0.8 and 1.6 Mbps. Wireless LAN Standards Although the HomeRF Working Group claimed some early market penetration of SWAP based products, by 2001, as SWAP 2.0 was being introduced with a 10 Mbps PHY layer data rate, the home networking market had been virtually monopolised by 802.11b products. The Working Group was finally disbanded in January 2003. HiperLAN/2 HiperLAN stands for high performance radio local area network and is a wireless LAN standard that was developed by the European Telecommunications Standards Institute's Broadband Radio Access Networks (BRAN) project. The HiperLAN/2 Global Forum was formed in September 1999 by Bosch, Ericsson, Nokia and others, as an open industry forum to promote HiperLAN/2 and ensure completion of the standard. The HiperLAN/2 PHY layer is very similar to the 802.11a PHY, using OFDM in the 5 GHz band to deliver a PHY layer data rate of up to 54 Mbps. The key difference between 802.11a and HiperLAN/2 is at the MAC layer where, instead of using CSMA/CA to control media access, HiperLAN/2 uses time division multiple access (TDMA). Aspects of these two access methods are compared in Table 6-17. The technical advantages of HiperLAN/2, namely QoS, European compatibility and higher MAC SAP data rate, have now to a large Table 6-17: CSMA/CA and TDMA Media Access Compared Media access method Characteristics CSMA/CA Contention based access, collisions or interference result in indefinite backoff. QoS to support synchronous (voice and video) traffic only introduced with 802. 11e. MAC efficiency reduced (54 Mbps at PHY = ca. 25 Mbps at MAC SAP). Dynamically assi

(Section "Spectrum Management at 5 GHz (802.11h), p. 160"). As a result, the previous support for HiperLAN/2 in the European industry has virtually disappeared. Given the overwhelming industry focus on products based on the 802.11 suite of standards, it seems unlikely that HiperLAN/2 will ever establish a foothold in the wireless LAN market, the clearest indication of this being perhaps that Google News returns zero hits for HiperLAN/2! Summary Since the ratification of 802.11b in July 1999, the 802.11 standard has established a dominant and seemingly unassailable position as the basis of WLAN technology. The various 802.11 specifications draw on a wide range of applicable techniques, such as the modulation and coding schemes shown in Table 6-18, and continue to motivate the further development and deployment of new technologies, such as MIMO radio and the coordination and control functions required for mesh networking. As future 802.11 Task Groups make a second pass through the alphabet, the further enhancement of WLAN capabilities will no doubt continue to present a rich and fascinating tapestry of technical developments. Chapter 6 has provided a grounding in the technical aspects and capabilities of current wireless LAN technologies. Chapter 7 now builds on that foundation in describing the practical considerations in implementing wireless LANs. This page intentionally left blank CHAPTER Implementing Wireless LANs There are many routes that lead from the identification of a user requirement for wireless networking to the operation and support of an installed WLAN, and the best approach to be taken will depend on the nature and scale of the project. In this chapter a five-step process is described that is scalable from a simple ad-hoc home network to a large scale corporate WLAN, linking multiple buildings. In small scale projects, such as in implementing a typical home or small office WLAN, some of these steps will be very short or may be eliminated altogether. Nevertheless, an awareness of the issues addresse

d in these steps will contribute to the successful implementation of even the smallest project. The five key steps in the planning and implementation of a wireless LAN area are as follows; 1. Evaluating requirements and choosing the right technology Establish the user requirements; what is it that the users want to be able to achieve and what are their 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 wireless LAN technologies rank against the technical requirements? Chapter Seven Selecting network hardware components; should a single or multi-vendor strategy be followed? What are the advantages and disadvantages? 2. Planning and designing the wireless LAN Surveying the RF environment; what other sources of RF energy or potential barriers to RF propagation are present in the target area of the wireless LAN? Designing the physical architecture; which architecture is right for the specific setting of the network? 3. Pilot testing Testing the chosen technology and architecture; does the chosen solution deliver the expected performance? 4. Implementation and configuration Putting the final wireless LAN in place and introducing it to the user group Configuring the appropriate security measures 5. Operation and support Keeping the wireless LAN operating efficiently and providing user support The following sections progress through each of these planning and implementation steps, while Chapter 8 is devoted to a more detailed description of the security aspects of WLANs. Evaluating Wireless LAN Requirements Establishing User Requirements If the wireless LAN is being implemented to support a large user group it will be important to gather a wide range of views on user requirements, perhaps by using a questionnaire or by interview. As a first step it may be necessary to raise awareness by demonstrating the technology to th

e prospective user group, SO that they are better able to give an informed view on requirements. Implementing Wireless LANs User requirements should be expressed in terms of the user experience rather than any particular solution or technical attribute, as they are independent of specific technologies. For example, in relation to performance expectations, a PHY layer data rate is a technical attribute, whereas the transfer time for a specified large file size is what the user is really concerned about. Common categories of user requirement are listed and discussed in Table 7-1. In virtually all aspects of user requirements the question of future proofing also arises; are future developments expected in the users' work processes, the type of technologies deployed in the users' business, growth in the business, etc., that will change the overall demands placed on the WLAN? Table 7-1: WLAN User Requirement Types Requirement type Considerations Usage model What user activities does the WLAN have to support? Are users routinely transferring large files over the network, such as Internet downloads or video editing? Is the WLAN required to support applications such as voice or video streaming, either now or in the future? Performance What are the user's performance expectations? If large data expectations files are commonly used, what are the required transfer times? Areal coverage What is the operating area in which users will need wireless network coverage? Do usage requirements vary at different locations within this area? Is future growth of the required coverage area expected? Mobility If users will move within the operating area while working, will they need to access the WLAN from several fixed locations (roaming) or will they need continuous service while in motion (mobility) - for example to support voice services? Device What types of user devices will need to connect to the interoperability network? User population What is the total number of users and user devices that are required to be supported? How many

users will typically require concurrent service? How much future growth is the network expected to cater for? Continued Chapter Seven Table 7-1: WLAN User Requirement Types - cont'd Requirement type Considerations Security How confidential is the information transferred across the network? What level of protection is required against unauthorised access? Battery life If mobile devices will be used in the network, how often will the user need to recharge battery operated devices? Economic What budget is available to implement the WLAN? Are there specific requirements that deliver high value and may justify higher cost solution? Table 7-2: WLAN Technical Attributes Requirement type Considerations Effective data rate The required data rate for a single user will be dictated by the usage model, for example by the typical file size and upload/download time, or by the requirements for voice or video streaming. As discussed in Chapter 6, effective data rates can be significantly lower than a standard PHY layer data rate, and will be further affected by adverse environmental factors such as RF interference. Network capacity What is the overall network capacity needed to provide the required level of service, given the current and future expected size of the user group and number of user devices? Required capacity will be a key factor both in the technology selection and in determining the appropriate physical architecture for the WLAN. Quality of service If the usage model includes applications such as VoWLAN, then guaranteed quality of service will be an important attribute to ensure performance expectations are met. Application support Are there specific technical attributes required to support particular usage models? Network topology What types of connections are required to meet user requirements? For example, peer-to-peer for local data sharing, point-to-point for linking buildings, etc. Security If users' confidentiality requirements are high, then data encryption, network access monitoring and other security meas

ures will be required. Implementing Wireless LANs Table 7-2: WLAN Technical Attributes - cont'd Requirement type Considerations Interference and If the WLAN will have to operate in an environment with coexistence other wireless networks, such as Bluetooth, or alongside cordless phones, then coexistence will need to be a consideration. Technology maturity Before standards have been agreed early products have an interoperability risk, while a fully mature technology may have limited scope for future development and risk early obsolescence as new usage models arise. The significance of this attribute will depend on whether the user requirements are within the proven capabilities of existing technology or require a leading edge solution. Operating range The required range will be determined by the physical extent and nature of the operating area, as well as the layout of components such as access points. The overall link budget will be important in implementing point-to-point connections (wireless bridges between buildings). Network scalability If the WLAN is likely to require more than a few access points, or significant future growth is anticipated, then ease of initial configuration and ongoing network management tasks will be a requirement, at least for the network manager. Establishing Technical Requirements Technical requirements follow from user requirements, by translating these into the specific technical attributes that are needed to deliver the user requirements (Table 7-2). For example, if there is a user requirement for rapid transfer of large files, for example for video editing applications, this will translate into a technical requirement for a high effective data rate. Some technical attributes, such as operating range and those relating to interference and coexistence, will be clarified following site surveying and initial planning of the physical layout of the network hardware. Evaluating Available Technologies Having established the technical attributes necessary to meet user requirements, the ava

ilable technologies can then be directly assessed against these attributes. A simple table, similar to the example shown Chapter Seven in Table 7-3, can be used to display the assessment, resulting in a transparent and objective comparison of the available solutions. More sophisticated evaluation methods can also be applied, for example, by assigning a weighting factor to each requirement and a score to each technical solution depending on the extent to which it meets the requirements. Network Capacity The total required network capacity will be determined by the sum of the bandwidth requirements of the maximum number of concurrent users expected on the network, with some allowance being made for the fact that this maximum will occur infrequently and some limited degradation of performance may be acceptable during brief periods of high usage. If this requirement exceeds the capacity of a single access point then multiple access points will be required, up to the limit imposed by the number of available non-overlapping channels. The total achievable network capacity for 802.11 networks defined by that limit is shown in Table 7-4. As described in the Section "Spectrum Management at 5 GHz (802.11h), p. 160", the 802.11h enhancements open up an additional 12 OFDM channels in the 5 GHz band, doubling the achievable network capacity for 802.11a networks. Operating Range The operating range of a wireless network link is influenced by a wide range of factors, from the modulation and coding scheme being used to the nature of the materials used in the construction of the building in which the network operates. The key factors are summarised in Table 7-5. The operating range for 802.11a/b/g networks for varying PHY layer data rates in a typical office environment is shown in Table 7-6, based on transmitted power levels of 100 mW for 802.11a/b, 30 mW for 802.11g and antenna gains of 2 dBi. Implementing Wireless LANs Table 7-3: WLAN Technologies; Technical Attribute Comparison Requirement type 802.11b 802.11g 802.11a 802.11n

PHY layer data rate 11 Mbps 54 Mbps 54 Mbps 200+ Mbps Effective data rate 6 Mbps 22 Mbps 25 Mbps 100 Mbps at MAC SAP (8-13 Mbps with 11b stations) Network capacity 3 non- 3 non- 12-24 non- 6-12 non- overlapping overlapping overlapping overlapping channels channels channels dual channels Quality of service Supported supported supported supported Interference and 2.4 GHz band 2.4 GHz band 5 GHz band 2.4 or 5 GHz coexistence Interoperable with 802.11b network Technology maturity Mature Mature Mature Immature Operating range Good indoor Good indoor Line of As for 11b range range including sight or 11a including wall wall penetration operation. depending penetration on frequency penetration Scalability Small number Small number Enterprise Enterprise of users per of users scale; many scale; many per AP users per AP users per AP Table 7-4: Effective Network Capacity Comparison for 802.11a, b and g Networks Network standard and operating mode MAC SAP Network rate (Mbps overlapping capacity per channel) channels (Mbps) 802.11b 802.11g (Mixed mode - RTS/CTS) 802.11g (Non-mixed mode) 802.11a 802.11a (With 802.11h enhancements) Chapter Seven Table 7-5: Factors Affecting WLAN Operating Range Factor Impact on operating range Frequency band As described in the Section "RF Signal Propagation and Losses, p. 112", free space loss is proportional to the logarithm of operating frequency, and increases by 6.7 dB with the increase in frequency from 2.4 to 5.8 GHz. Transmitter power These two factors are grouped together since they determine and receiver the end points of the link budget, as described in the Section sensitivity "Link Budget, p. 116" Modulation and Higher data rate modulation and coding schemes are less coding scheme robust in that they require correspondingly higher received signal strength to assure accurate decoding. Other things being in equal range therefore decreases for higher data rates. Environmental Construction materials, particularly metal objects, have a factors major influence on path loss if RF signals ha

ve to pass through walls, ceilings, floors or other obstructions. Path loss is also highly frequency dependent and for all practical purposes a line-of-sight is required for communication in the 5 GHz band. Table 7-6: WLAN Indicative Indoor Range vs. PHY Data Rate PHY data rate (Mbps) 802.11a 802.11b 802.11g Implementing Wireless LANs Selecting the Technical Solution With the demise of competing standards such as HiperLAN/2 and HomeRF, the essence of the technical choice is simply - which 802.11 flavour best fits the bill? While the requirements analysis may point to a clear winner among the available technical options, if a selection needs to be made between operating in the 2.4 or 5 GHz bands then an RF site survey, described in the next section, should be conducted as an input to that decision. Similarly, if the requirements dictate that network capacity needs to be stretched beyond the throughput of a single channel, for example with multiple access points fully exploiting non-overlapping channels, then an initial physical layout may need to be made for both 2.4 and 5 GHz options. Some on-site physical testing may also be valuable prior to making the final decision, for example to confirm the achievable range if a 5 GHz network is envisaged in a constricted indoor environment. Future hardware developments may soon make the choice between 2.4 and 5 GHz operation irrelevant. As increasing volumes of dual band radios are shipped, supporting both 802.11g/b and 802.11a, and prices fall to parity with single band products, it will be cost-effective to implement a dual band WLAN that makes the best use of the characteristics of both RF bands. Planning and Designing the Wireless LAN Surveying the RF environment Conducting a site survey is an important step in planning and designing all but the simplest WLAN. It is important to determine the impact of environmental factors on radio wave propagation in the operating area of the LAN, and also to test for the presence of RF signals that will interfere with WLAN performan

ce. The objective of the site survey is to gather enough information to plan the number and location of access points to provide the required coverage, in terms of achieving a minimum required data rate over the operating area. Chapter Seven There are two types of site survey that can be performed - a noise and interference survey and a propagation and signal strength survey. Noise and Interference Survey This survey looks specifically for the presence of radio interference coming from other sources, such as nearby networks, military installations, etc., that could degrade WLAN performance. The main aspects of this type of survey are described below in Table 7-7. Propagation and Signal Strength Survey A well executed propagation and signal strength survey will help to ensure that network resources are correctly located SO that the planned network will not suffer from coverage holes, resulting in areas of poor network performance, and will also ensure that network capacity is properly planned. The main aspects of this type of survey are described below in Table 7-8. Table 7-7: Noise and Interference Survey Survey aspect Description Objective Assess the noise floor (RF power per unit of bandwidth, dBm/MHz) across the intended bandwidth to be used by the WLAN. Identify the distribution of RF energy within the bandwidth (frequency, continuous or sporadic, peak and average power levels). Equipment used Stand-alone RF spectrum analyser or PC equipped with a wireless interface card and spectrum analysis software. Survey Site walk-through test prior to access point placement. technique Extended point measurements should be made at selected locations to identify any intermittent interference sources. Application of Noise floor measurements will be used in the link budget survey results calculation to assess the effective range of an RF link given the hardware specifications (transmit power, receiver sensitivity, antenna gains). Interference results will indicate any limitations on bandwidth usage, such as channels that sh

ould be avoided, and in extreme cases may dictate a choice between 2.4 and 5 GHz operation. Implementing Wireless LANs Table 7-8: Propagation and Signal Strength Survey Survey aspect Description Objective Identify the coverage pattern, received signal strength and achievable data rate for given access point and client station locations within the WLAN operating area. Equipment used Laptop or handheld PC equipped with wireless interface card and site survey software. Ideally the survey receiver hardware should be identical to the planned client station hardware (same wireless NIC), otherwise allowances will need to be made, for example for differing receiver sensitivity or antenna gain. Combining the survey tool with a GPS navigation module can help in transferring measurements onto site plans. Survey Site walk-through test following preliminary access point technique placement, measuring received signal strength and maximum data rate at each location. If both 2.4 and 5 GHz operation are under consideration, the survey will need to be conducted in each band, as propagation patters will differ widely between these bands. Application of Results will show how the ideal omnidirectional propagation survey results pattern is affected by the operating environment (office partitions, walls, cabinets, lift shafts, etc.). Achievable data rate versus range will dictate the required placement of access points as input to planning the physical hardware layout. Figure 7-1 shows a typical display of propagation and signal strength survey results, which can be used to identify areas where RF propagation is adversely affected by local environmental conditions. Combined with a similar graphical display of noise floor measurements this will give an indication of potential low SNR areas, which may need to be addressed by adjusting access point or antenna locations or by changing transmitter power settings or antenna gain. Designing the Physical Architecture Having built up a picture of the RF environment and gathered data on propagat

ion and signal strength in the operating area, a provisional physical layout of the WLAN can be created. The objective of this stage is to establish a layout of hardware that will ensure complete RF coverage and deliver the required bandwidth to wireless client stations. Chapter Seven R Live File Edit View Actions Help Map Type AP-Coverage Caverage Rate : Layered SNR Channel All Merged Alpha Alpha HQ Floor1 AP Count 18 AM Count :6 Grid HQ Floors Floori Main Campus 1.127457 4991573 1161.a 11/1/157x 11.1.15/M63.a. 1.1.43/161.a 1.1.44/163a 21/157a 1:16:15 1-4 247 Areas Suggested APs/AMs Deployed APs Bottom Right Details 133,44 Figure 7-1: Display of Propagation and Signal Strength Survey Results (Courtesy of Aruba Wireless Networks Inc.) Network Physical Layout Design Planning the physical architecture starts with the floor plans of the operating area and the results of the propagation and signal strength survey, and results in a layout plan detailing; the required number of access points an optimal antenna type and location non-conflicting operating channel proper power setting for each access point. Factors influencing physical layout are described in Table 7-9. The channel allocation patterns shown in Figure 7-2 are based on the three non-overlapping channels 1, 6 and 11 in the 2.4 GHz band. However, in those regulatory domains that permit 13 channels in this band (Europe, Japan, etc., as discussed in the Section "802.11b PHY Layer, Implementing Wireless LANs Table 7-9: Factors Influencing WLAN Physical Architecture Parameter Factors influencing the physical architecture Number A preliminary count can be established by dividing the overall operating area per access point by the coverage area determined from the propagation and signal strength survey. The area should be taken out to the contour at which the required data rate could be maintained. The effective coverage area of an access point will be reduced if the propagation pattern is far from omnidirectional as a result of nearby obstacles. Optimal antenna The

optimal location for an access point with an location omnidirectional antenna will generally be close to the centre of the area to be covered, in a position that maximises line-of-sight to client stations and is clear of obstructions, particularly metal objects such as filing cabinets. An elevated location can be very effective, for example, a ceiling mounted unit. Operating Any channel that shows significant background or sporadic channel noise should be avoided. The available non-overlapping channels can then be allocated to access points based on their initial locations. A typical channel allocation pattern for the three non-overlapping channels in the 2.4 GHz band is shown in Figure 7-2. Power setting In general, the number of access points will be minimised if maximum permitted transmit power is used. Reasons to adopt a lower power setting may be to reduce out of building propagation or to avoid interference with other RF systems. Conversely high power settings may be required to combat local conditions of high RF noise or high path loss. p. 152"), it is possible to operate four access points with minimal frequency overlap on channels 1, 5, 9 and 13, potentially increasing network capacity by one-third. This would permit the additional channel allocation patterns shown in Figure 7-3. Early WLAN layouts were often developed on a trial-and-error basis, by temporarily deploying access points, taking signal strength and throughput measurement (essentially repeating a site survey) and then relocating or adding access points to fill in any identified holes in RF coverage. Planning tools such as Wireless Valley's "LAN Planner" are available that improve WLAN design by simulation and graphical display of the Chapter Seven Grid pattern Cellular pattern Figure 7-2: Channel Allocation Pattern for 802.11b Access Points with Three Non-overlapping Channels expected performance of the network. Information such as received signal strength indicator (RSSI), signal to interference ratio (SIR), signal to noise ratio (SNR), th

roughput and bit error rate (BER) can be displayed on digitised site plans, allowing more accurate planning in complex environments. Besides automated placement and configuration of network components, these tools can also ease the later stages of implementation by automatically generating bills of materials and maintenance records. 4-channel grid pattern Cellular pattern Figure 7-3: Channel Allocation Patterns for Four Non-overlapping Channels Implementing Wireless LANs Design and Deployment Using Wireless Switches Wireless switches, described in the Section "Wireless LAN Switches or Controllers, p. 48", not only change the way in which network traffic is routed between stations, but also provide in-built tools that aid the design and later the configuration process, particularly in large-scale WLAN installations. Typically a wireless switch tool kit will include; Automatic layout planning for capacity and coverage (Table 7-10) One-click configuration of multiple access points and WLAN switches Simple monitoring and operation, from the detection and location of rogue access points to automatic transmitter power adjustment to eliminate coverage gaps and optimise network performance. Network Bridging - Point-to-point Wireless Links If the WLAN requirements include linking two separate operating areas, for example providing a link from a WLAN in one building to a wired or wireless LAN in a second, then a point-to-point link will need to be designed. Since RF propagation is more predictable outdoors than indoors, this will generally be more straightforward than designing indoor layouts as described above. Provided suitable antenna locations can be identified that offer a direct line-of-sight between the locations, a link budget calculation can be performed as described in the Section "Link Budget, p. 116". Table 7-10: Wireless Switch Automated Planning Tools Wireless switch feature Description Automated site survey Most wireless switches provide automated site survey tools tools that generate input survey data for s

imulation of the WLAN operating environment. Access point location Tools allow import of building blueprints and construction planning specifications from CAD programs and in-built propagation models are used to determine optimal access point location, taking account of construction materials, path losses and attenuation. Chapter Seven The required combination of transmit power and antenna gain needed to deliver sufficient received signal strength to achieve the desired data rate can be determined, given the link range and intended RF band. Pilot Testing The design process described above will have established a layout of WLAN hardware which aims to deliver the required data throughput with complete RF coverage of the operating area. A pilot test of the design solution, prior to complete installation, will be helpful to ensure that no requirements have been missed and no limitations of the selected technology have been overlooked. Aspects of pilot testing are summarised in Table 7-11. A pilot test will involve the installation of a number of access points of the same type as are intended for the final installation, to cover part of the operating area. Selecting the part of the operating area that posed the greatest difficulty in the design stage - either because of interference or identified propagation issues - will provide the most robust challenge to the design solution. The pilot test should include monitoring and user responses to the day-to-day performance of the pilot installation, as well as stress testing - running under extreme load conditions. The results of the pilot test will either validate the installation proposed at the design stage, or they may indicate that adjustments in the design are necessary if user requirements and performance expectations are to be fully met. A repeat pilot testing may be valuable if major changes in the design solution are indicated. Installation and Configuration WLAN Installation Installation should follow a systematic approach in which each building block of the WLAN

is installed and tested before moving on to the next. This will ensure that any problems are identified as early as possible - once the whole network is in place it may be much more difficult to identify a single faulty element. Implementing Wireless LANs Table 7-11: Aspects of WLAN Pilot Testing Pilot test feature Description Stress testing Stress testing of the pilot installation will involve loading the wireless LAN with the most demanding data transfer requirements, such as video or voice streaming or transfers of large files. Gradually increasing the number of concurrent users of high rate applications will test the achievable throughput limit. Maintain and Access point logs should be enabled during the pilot test analyse period, and analysed to show peak usage times and identify user logs the types of services that were used during the pilot. A comparison with the stated user requirements will test whether the expected usage patterns are realistic. Conduct post-pilot User surveys conducted after the pilot has been in operation user surveys for some time can highlight the type and frequency of any problems, and test whether user expectations are being met. Problem reports can be matched to access point logs to highlight any bottlenecks in the pilot installation that may point to necessary design changes before full-scale implementation. The key implementation steps are described in Table 7-12. If the installation involves multiple access points, follow these steps for each BSS in turn - verifying operation of each before moving on to the next. After installing the first access point, Steps 4-6 can be followed for each station to be connected to the access point. Steps 1-6 can then be followed for the second and any additional access points. WLAN Configuration The initial configuration will ensure that WLAN parameters are set-up to enable communication between the installed access points and stations, and ensure the coexistence of adjacent BSSs. After completing the configuration of access points and station

s, the network operating system may also need to be configured. Chapter Seven Table 7-12: WLAN Implementation Steps Implementation step Description/Considerations (1) Install Ethernet If access point locations have not been pilot tested it cabling to the may be desirable to lay temporary cabling until the planned access access point locations have been verified in practice, point location especially if cable laying is expensive and disruptive. Test the cabling by connecting a laptop to the wired network via the new cable and "ping" testing connectivity (see the Section "Analysing Wireless LAN Problems, p. 241"). (2) Install the first Follow the vendor's instructions and use any mounting access point at its kit provided, for example, for wall, ceiling or above planned location and suspended ceilings. Connect network cables and connect to the wired antenna(s), following vendor recommendations regarding network antenna orientation. Connect power cable unless power- over-Ethernet is being used. (3) Configure the Configuration is described further in the following access point settings section. Depending on the hardware in use, it may be necessary to install configuration software on a computer connected to the wired network, or to configure the access point via a web based utility. (4) Install wireless Check vendors instruction to determine the correct NICs in the installation sequence (drivers first or hardware first), computers that will which is often operating-system dependent (e.g. connect via this hardware first for Windows XP). access point (5) Configure a After installation of software drivers, the wireless wireless network NIC may need to be identified to the computer's connection for each operating system as a new network connection. new station (6) Verify operation Confirm network connectivity for stations connected of all stations through the first access point. Access Point Configuration The details of access point configuration will depend on the specific hardware selected, but Table 7-13 gives a summar

y of the basic configuration parameters that will be applicable in all cases. If the access point is a dual- or tri-mode device, with multiple radios, each will require separate configuration. Some access points may also allow other PHY layer parameters to be configured, such as the fragmentation Implementing Wireless LANs Table 7-13: Access Point Configuration Parameters Parameter Configuration considerations IP address (plus The access point may have a default IP address, subnet mask and defined by the manufacturer, that will enable a direct default gateway) connection through a web browser. Alternatively an IP address may be assigned to the access point by a DHCP server in the wired network or a static IP address (plus subnet mask and default gateway) may be assigned by connecting a PC to the access point's configuration port. The service set ID should be changed from its default value as a basic security measure in every network. A policy for SSID assignment may be defined for large WLAN installations. SSID broadcast SSID broadcast within beacon frames can be either enabled or disabled. Disabling SSID broadcast is a further security measure, discussed in Chapter 8. Maximum transmit Setting the maximum allowable transmit power level power to comply with local regulations. Radio channel Selection of the operating channel within the range allowed by local regulations. Some access points may include an automatic search for the least congested channel. Operating mode In the case of 802.11g networks, a selection can be made between mixed mode and g-only mode, depending on whether 802.11b stations will also be operating in the network. Security Selection of security modes (64-bit WEP, 128-bit WEP, WPA-PSK, etc.) and entry of passphrase or encryption keys, transmit key and authentication mode (described further in Chapter 8). Antenna configuration An access point with multiple antennas may be configured to use one specified antenna, or to use all antennas in diversity mode - selecting the antenna that gives the stron

gest signal. Chapter Seven D-Link D-Link AirPlus AirPlus 802.11g/2.4GHz Wireless Router 802.11g/2.4GHz Wireless Router Advanced Tools Status Advanced Tools Status DI-524 DI-524 Wireless Settings Wireless Performance These are the settings for the AP(Access Point portion These are the Wireless features for the AP(Access Point portion Network ID(SSID) Roslyn Interval (msec, range -1000 default 100) Channel Wizard Threshold (range 256-2432 default 2432) Security (range 256-2346 default 2346, even number Wireless WEP Encryption Key Mode (range 1-65535 default 3) C31800CC20 mixed mode mode Filter (Mbps) Open System o Shared Key Both Firewall Enable Disable Input 10 HEX characters (HEX) 0-9,A-F, Apply Cancel Help Apply Cancel Help Performance Figure 7-4: Typical Access Point Configuration Screens limit or the maximum number of retries before a packet is dropped. Figure 7-4 shows typical set-up screens for configuring access point operating parameters. Wireless Station Configuration As for access points, station configuration details will be hardware dependent, but Table 7-14 describes the basis wireless NIC configuration Table 7-14: Wireless NIC Configuration Parameter Configuration considerations Network mode Select infrastructure mode if the station will be associating with an access point, or ad-hoc mode if the station will only be connecting to other stations and not to an access point. Enter the SSID of the access point that this station will be connecting to. Radio channel Selection of the operating channel within the range allowed by local regulations. Operating mode As for access points, for 802.11g hardware a selection can be made between mixed mode and g-only mode in 802.11g networks. Security Security settings (passphrase or encryption keys, transmit key and authentication mode) must match those entered for the access point that the station will be associating with. Implementing Wireless LANs D- ink AirPlus G Wireless Utility D-1 ink AirPlus G Wireless Utility Roslyn Link Info Link Info Wireless Mode Infrast

ructure Adhoc Channel Channel 6 Configuration Authentication Configuration Profile IP Settings Disable Data Encryption Advanced 64 bits (40+24) 10 Hexadecimal di Advanced Key Length Power Mode F Fast Save IEEE802 1X Disabled Site Survey Site Survey Launch Utility on Startup Enable C20B30CC40 About About Apply Data Packet Parameter Authentication Config Fragmentation Threshold Settings RTS Threshold Apply Figure 7-5: Typical Wireless NIC Configuration Screens parameters that will be applicable in all cases. Figure 7-5 shows typical set-up screens for configuring wireless NIC operating parameters. Network Operating System Configuration If the WLAN is an extension of an existing wired network, then no additional network operating system (NOS) configuration will be required. However, if the WLAN is creating a new network, a number of NOS configuration tasks will also have to be completed. Details will depend on the specific NOS in use, but typical tasks will include; ensuring that network protocols such as TCP/IP are installed ensuring that networking software is installed e.g. for file and printer sharing identifying workgroups of users that share resources enabling parts of the network file system for common or workgroup access enabling devices such as printers, scanners, etc., for shared access. Chapter Seven Automatic WLAN configuration and management A traditional WLAN deployment, based on first generation or "fat" access points, will have limited or no built-in network management capabilities, and initial configuration and ongoing management will generally be performed using a web-based user interface. Network management tasks, such as a change of security settings, RF operating channel, transmit power level or access policy, will have to be implemented individually on each access point. For a corporate WLAN, as the number of access points grows, this will be very time consuming and will quickly become unmanageable. Second generation WLAN hardware, in the form of wireless switches, are designed to enable these

to the operation and support of the WLAN. Initial user training, day-to- day help desk support and hardware maintenance will need to be planned and provided, depending on the size of the user group. Two key tasks to assure that the installation continues to meet user requirements are network performance monitoring and the control of future changes in the network installation or configuration. Network Performance Monitoring For medium to large-scale WLAN implementations, the network manager will need to keep an overview of network performance in order to identify and diagnose the nature and location of any problems, and quickly determine a solution. A variety of WLAN management tools are available to ease this performance monitoring task. Typically a graphical interface, based on building blueprints or other plans of the operating area, allows the network manager to view performance data collected from access points and interface cards. This real time performance data, collected using SNMP queries to access points and stations, as well as from system software, will be able to identify; Active access points and client stations Average data rates, retry rates and overall network utilisation Noise levels and interference by area Network areas or individual access points experiencing significant error rates. Aruba Network's RF LiveTM is an example of this type of WLAN management software. A view of the real time management dashboard is shown in Figure 7-6. Network Change Control Having set-up the WLAN based on a well thought out physical layout and configuration plan, future changes must be given the same degree of Chapter Seven AirMagnet Demo 2 2.7% File View Tools V All MAC Address Preomble PCF/DCF First 00:03:2f:03:18:02 weec30wendriver 15:36:16 15:36 Agere 50.91 INTERMEC 15:36:16 15:3E 00:01:F4:ED:70:CC woec30enteracys 15:36:16 15:36 00:50 DA:00 8C 67 netops 15:36:15 15:36 SymbolA1:D1:C4 awsec30symbol 15:36:15 15:36 Akonet:55:10:c9 wsec30cis.co Short 15:36:15 15:36 vernier 15:36:15 15:36 Agere 0F:50:04 weec? Inone

p 15:36:15 15:36 SymbolA1:D4E1 dabswlan 15:36:15 15:3E Avonet:57:90.88 weec20weep Short 15:38:15 15:38 00:06:25:53:0c:08 15:36:15 15:36 SymbolA1 D4E5 nocwinn 15:39:15 15:36 LucentFDC8:C7 15:36:15 15:36 175:61:37:63 303com 15:36:15 15:36 SymbolA28F61 hdwlan 15:36:15 15:38 SymbolA20F66 noowlan 15:36:15 15:30 SSID (18) Nokia D4 an labowsec 15:38:15 15:38 AdHoc Symbol 63 Symbol 15:36:15 15.3E Intrastructure P (18) STA(G) Expert Advice Performance (4.0.18) Security (28.3.0) Frame Address Type Unicast (30%) Muticast(11 Broadcast (69%) Broadcast Multicard Total Frames Scan Stopped Start Channel Infrastructure AirWISE Charts Decodes Tools Figure 7-6: WLAN Management Software Display (Courtesy of AirMagnet Inc. © 2006 AirMagnet R is a registered trademark.) thought and planning if the effectiveness of the network is to be maintained. The addition of an access point, perhaps aimed at filling in a coverage gap or providing additional network capacity for a new concentration of users, could easily have the opposite effect on performance if the new hardware is not configured for coexistence with the existing set-up, with appropriate channel selection, transmit power and security settings. A documented set of network policies and a supporting change control procedure will ensure that future change is managed in such a way that performance is maintained or enhanced. Network policies may address issues such as; Supported standards (e.g. 802.11b/g) Supported hardware vendors or NIC models SSID management Security and encryption requirements (e.g. personal firewalls, WEP/WPA). Implementing Wireless LANs A change control procedure should define how proposed changes to the installation, including hardware, software and configuration settings, will be technically reviewed and endorsed prior to implementation. The procedure should cover; Scope of the procedure - what types of changes are controlled Process and documentation required for proposing changes Roles and responsibilities of reviewers and decision makers Authorities required

to endorse different types of changes. A Case Study: Voice over WLAN In order to meet the demands of supporting voice applications, a wireless LAN needs to fulfil stringent performance requirements, particularly if the provision of voice services was not a consideration during the initial design. Close attention will need to be paid to wireless coverage, quality of service and seamless roaming, recognising that the roaming area for voice coverage is likely to be different from the operating area for other roaming coverage such as laptop connections. VoWLAN Bandwidth Requirements Voice usage puts high demands on WLAN bandwidth, and an estimate of the number of concurrent calls that need to be supported will be a key factor in determining either the required design or the suitability of an existing WLAN design. Although a single voice session requires around 64 kbps of bandwidth, or as little as 10 kbps with compression, IP and 802.11 MAC protocol overheads increase the required bandwidth to around 200 kbps per session. Collisions on the shared wireless medium will further limit the number of concurrent voice sessions. The actual number of concurrent calls that can be supported by a single access point, while providing voice quality comparable to a toll call, will depend on the wireless standard in use, as shown in Table 7-16. The concurrent call limits noted here are indicative only, and apply to a network carrying voice traffic alone. If the network is carrying data Chapter Seven Table 7-16: VoWLAN Capacity of 802.11 WLANs Standard PHY data rate (Mbps) MAC SAP data rate (Mbps) Maximum concurrent voice calls 802.11b 802.11b + g 802.11g ca. 20 802.11a ca. 25 as well as voice traffic, call quality will be affected and the maximum number of concurrent calls will drop. As these capacity limits are approached, load balancing will be required to ensure that access points do not become overloaded. Access points alone will not have the overview of the available infrastructure or total traffic to perform this function, but

a wireless switch will be able to monitor the total number of voice sessions being routed at any moment and intelligently manage capacity by handing off VoWLAN traffic where possible to alternative access points. RF Coverage Voice services put greater demands on wireless coverage, SO that roaming users do not encounter coverage gaps in corridors or stair wells while talking on their VoWLAN phones. If the WLAN has been designed to provide coverage of work areas and meeting rooms, a more focussed propagation and signal strength survey will be required when planning for VoWLAN service. Quality of service to voice clients will be improved if only the maximum supported data rate is enabled on all access points. This ensures that overall network throughput is not slowed when a client moves away from the access point, but also means that a higher density of access points will be necessary to give adequate RF coverage to support voice traffic. Implementing Wireless LANs Quality of Service VoWLAN requires guaranteed quality of service in order to minimise packet loss, delay and jitter that cause degradation of voice quality. As described in the Section "Quality of Service (802.11e specification), p. 157", the 802.11e standard has been developed to address the lack of QoS in the original 802.11 specification, and the Wi-Fi Alliance has also released wireless multimedia (WMM) as an interim subset of the 802.11e specification. WMM and 802.11e prioritise voice traffic over video, best effort and background traffic categories, and 802.11e or WMM compliance will be an essential technical requirement for WLAN hardware in all but the least demanding VoWLAN applications. However, 802.11e and WMM define traffic categories by device and not by application, SO that frames transmitted by a laptop user running a software phone will be queued according to the device priority, which will typically be best effort. Future enhancements of QoS standards are likely to provide for prioritisation by application. Seamless Roaming Complete RF co

verage alone is not sufficient to ensure the seamless roaming required for uninterrupted VoWLAN service. Roaming VoWLAN clients also need to make fast transitions between access points, avoiding latency and packet loss during hand-off, SO that service quality is not degraded or interrupted. When a voice client transitions from one access point to another, association and authentication at the new access point must be fast to avoid degradation of voice quality. Typically, a VoWLAN performs optimally with packet delays and jitter of under 50 ms, and a call will drop out when delays approach 150 ms. By comparison, associating and authentication with an access point typically takes from 150 to 500 ms, and measured roaming times for VoWLAN clients can be anything from 1 to 4 seconds with multiple clients making concurrent transitions. To achieve the faster transitions required for seamless roaming, pre-authentication must be completed prior to hand-off, as provided for in the 802.11r standard. Chapter Seven Dual-mode handsets are now available that allow roaming between cellular phone networks and WLANS, with the access point or wireless switch controlling the hand-off between the VoWLAN service and a cellular phone service. Pilot Testing As described in Chapter 6, a number of upcoming 802.11 enhancements are aimed at improving RF management, QoS and roaming in order to improve the capability of 802.11 networks to support VoWLAN services. Until these standards are ratified and published, and compliant products become available, VoWLAN deployments will require careful pilot testing to ensure that the required service can be delivered. Stress testing should be conducted by loading the WLAN with multiple concurrent calls and testing voice quality with and without background data streams being present on the network. Coverage and roaming ability can also be assessed using modelling tools, but ultimately the quality of voice services can be best confirmed by pilot testing. VoWLAN Security When using a conventional telephon

e system, physical access to telephone lines or to a private branch exchange (PBX) in an office is required in order to intercept voice traffic. The physical security of these systems means that the encryption of voice traffic over a conventional telephone system is only justified for highly security-sensitive organisations. In the case of VoWLAN, voice traffic going outside the organisation will pass over the completely unsecured Internet, and it becomes relevant for many more organisations to provide the same degree of security for voice for data traffic. VoWLAN services will be susceptible to the range of security attacks described in Chapter 8, and will be particularly sensitive to Denial of Service attacks in view of the delay sensitive nature of voice traffic. Specific VoWLAN/VoIP security considerations are discussed in the Section "VOWLAN and VoIP Security, p. 239". A further security concern for VoWLAN/VoIP services is the possibility of unsolicited bulk messages being broadcast to Internet connected Implementing Wireless LANs phones. As IP telephony becomes more widespread, SO called spam over internet telephony (SPIT) could become as pervasive as spam e-mail. Technology to counter this threat is under development, based either on filtering and deleting unwanted calls based on characteristics such as call frequency and duration, or identifying authentic calls by enabling these to be digitally signed and checking signatures against an authorised caller list. This page intentionally left blank CHAPTER Wireless LAN Security The Hacking Threat The flexibility of wireless networking is bought at the price of an increased need to think about security. Unlike a wired network, where signals are effectively limited to the connecting cables, WLAN transmissions can propagate well beyond the intended operating area of the network, into adjacent public spaces or nearby buildings. With an appropriate receiver, data transmitted over the WLAN may be accessible to anyone within range of the transmitter. Although the spr

ead spectrum modulation technologies described in Chapter 4 were originally conceived for military applications where jamming was the main concern, any device conforming to a wireless standard such as 802.11b will be able to intercept wireless data traffic from a WLAN operating that standard. Additional security measures must be taken to prevent unauthorised access to user data and network resources. The easy accessibility of wireless networks resulted in a new industry springing up among the early adopters to map out networks and make others aware of the opportunity for free, although often illegal, access. War Driving and War Walking describe the pursuits of driving or walking around with a wireless enabled laptop or handheld device and seeking out wireless networks. War Chalking sprang up in London in 2002 as a way of making others aware of nearby wireless networks that could be freely accessed. Some WLANs are deliberately left without security measures enabled in order to allow Chapter Eight Open Node Closed Node Bandwidth Access contact WEP Node Bandwidth Figure 8-1: War Chalking; Table of Symbol Elements and Captions free public access, and War Chalking symbols were intended to help people living in or visiting a city to identify these networks in order to connect to the Internet (Figure 8-1). The site www.warchalking.org gives more information, as well as an interesting discussion on the legal and moral aspects of accessing unsecured WLANs. These free access points are of no interest to the determined hacker (or cracker), who is motivated by the challenge of breaking into a secured network. The measures described later in this chapter should be followed to ensure that access to a wireless network is as secure as it can be, and to minimise this risk of unauthorised access and use. At home, it is not just deliberate hackers who might see an unsecured wireless network as a free resource. If basic security measures are not enabled, any wireless enabled computer in a neighbouring house or apartment will be able

to connect to the network and make free use of resources such as an Internet connection. Wireless LAN Security Threats The type of security threats faced by a wireless LAN are many and varied, and although initially targeted at the PHY and MAC layers, the ultimate goal is to access or disrupt data or activity at the application layer. A few of the main vulnerabilities are described below. Denial of service (DoS) attacks - an attacker floods a network device with excessive traffic, preventing or seriously slowing Wireless LAN Security normal access. This can be targeted at several levels, for example, flooding a web server with page requests or an access point with association or authentication requests. Jamming - a form of DoS in which an attacker floods the RF band with interference, causing WLAN communication to grind to a halt. In the 2.4 GHz band this could be done using Bluetooth devices, some cordless phones or a microwave oven! Insertion attacks - an attacker is able to connect an unauthorised client station to an access point, either because no authorisation check was made or because the attacker masqueraded as an authorised user. Replay attack - an attacker intercepts network traffic, such as a password, and uses it at a later time to gain unauthorised access to the network. Broadcast monitoring - in a poorly configured network if an access point is connected to a hub rather than a switch, the hub will broadcast data packets that may not be intended for wireless stations, and these can be intercepted by an attacker. ARP Spoofing (or ARP cache poisoning) - an attacker can trick the network into routing sensitive data to the attacker's wireless station, by accessing and corrupting the ARP cache in which MAC and IP address pairs are stored. Session hijacking (or man-in-the-middle attack) - a type of ARP spoofing attack in which an attacker breaks a station's connection with the access point, by posing as the station and disassociating itself, and then poses as the access point to get the station to associa

te with the attacker. Rogue access point (or evil twin intercept) - an attacker installs an unauthorised access point with the correct SSID (the twin). If the signal is strengthened using an amplifier or high gain antenna, clients stations will preferentially associate with the rogue access point and sensitive data will be compromised. Cryptoanalytic attacks - an attack in which the attacker uses a theoretical weakness to break the cryptographic system. An example is the weakness of the RC4 cipher that leads to the Chapter Eight vulnerability in WEP (see the Section "WEP Wired Equivalent, Privacy Encryption, p. 209"). Side channel attacks - an attack in which the attacker uses physical information, such as power consumption, timing information or acoustic or electromagnetic emissions to gain information about the cryptographic system. Analysis of this information might allow the attacker to determine an encryption key directly or a plaintext message from which the key can be computed. Although the range of threats is wide and varied, in most cases executing these kinds of attack requires a high level of technical expertise on the part of the hacker. The risk to network security can be significantly reduced by enabling the full range of available security measures described in the following sections. WLAN Security Table 8-1 summarises the range of generic security measures that have been developed to protect wireless LANs from the threats and vulnerabilities described above. As described in the Section "WEP - Wired Equivalent Privacy Encryption, p. 209", the original 802.11 standard included only limited Table 8-1: Wireless LAN Security Measures Security measure Description User authentication Confirms that users who attempt to gain access to the network are who they say they are. User access control Allows access to the network only to those authenticated users who are permitted access. Data privacy Ensures that data transmitted over the network is protected by encryption from eavesdropping or any other unauthori

sed access. Key management Creation, protection and distribution of keys used for encrypting data and other messages. Message integrity Checks that a message has not been modified during transmission. Wireless LAN Security authentication and weak encryption. The interim development and deployment of enhancements to 802.11 security has been led by the Wi-Fi Alliance with the release of WPA and WPA2. The shortcomings of the original 802.11 standard were addressed with the ratification in 2004 of the 802.11i standard, which provides a standards basis for WPA and WPA2. The progressive enhancements to WLAN security, and the technologies underlying these enhancements, are described in the Sections "WEP - Wired Equivalent Privacy Encryption, p. 209", "Wi-Fi Protected Access; WPA, p. 212" and "IEEE 802.11i and WPA2, p. 219", and the Sections "WLAN Security Measures, p. 230" and "Wireless Hotspot Security, p. 236" then return to the practical aspects of ensuring security in wireless LANs, at wireless hotspots and some specific aspects of VoWLAN security. WEP - Wired Equivalent Privacy Encryption As the name suggests, the intention of WEP was to provide a level of security equivalent to a wired network, although this aspiration was not achieved because of a fundamental cryptographic weakness. From the list of security measures summarised in Table 8-1, WEP provides a limited degree of access control and data privacy using a secret key, typically a passphrase, that is entered into the access point and is required to be known by any station attempting to associate with the access point. Without knowledge of the passphrase, a station will be able to see network traffic but will not be able to associate or easily decrypt data. WEP encryption translates the passphrase into a 40-bit secret key, to which a 24-bit initialisation vector (IV; see Glossary) is added to create a 64-bit encryption key. As an interim attempt to strengthen WEP encryption, some vendors enhanced the key length to 128-bit (104-bit + 24-bit IV). This proved t

o be largely a cosmetic enhancement since, as described below, the underlying vulnerability meant that an eavesdropper could still derive the key by analysing roughly 4 million transmitted frames, whether 40-bit or 104-bit keys were used. The input data stream, known in cryptographic terminology as the plaintext, is combined with a pseudo-random key bit stream in an XOR (exclusive OR) operation to create the encrypted ciphertext. WEP creates the key bit stream using the RC4 algorithm to make a pseudo-random Chapter Eight Input counters S[i]+S[j] Output key byte S[i] + S[j] Figure 8-2: Key Stream Generation Using the RC4 Algorithm selection of bytes from a sequence S that is a permutation of all 256 possible bytes. As shown in Figure 8-2, RC4 selects the next byte in the key stream by; Step (1) incrementing the value of a counter i, Step (2) incrementing a second counter, j, by adding the value of S(i), the i'th byte in the sequence, to the previous value of j, Step (3) looking up the values of the two bytes S(i) and S(j), indexed by the two counters, and adding them together modulo 256, Step (4) outputting the byte K indexed by S(i) + S(j), i.e. K=S(S(i) - + S(j)). The values of bytes S(i) and S(j) are then exchanged before returning to step (1) to select the next byte in the key stream. The initial permutation of bytes in S is determined by a key scheduling algorithm which uses a similar manipulation of bytes within S, starting with the identity permutation of bytes (0, 1, 2, 3, 4, ..., 255), but at each Step when the counter j is incremented, a byte from the 64-bit or 128-bit encryption key is also added to the counter. WEP also provides limited message integrity checking using a cyclic redundancy check (CRC-32; see Glossary) to compute a 32-bit integrity check value (ICV) which is appended to the data block before encryption. The full WEP computational sequence is then as follows (see Figure 8-3); Step (1) the ICV is computed for the data block to be transmitted in the frame Wireless LAN Security Plaintext Pla

intext Initialisation Secret Keystream vector Ciphertext Figure 8-3: WEP Encryption Process Step (2) the ICV is appended to the data block Step (3) The initialisation vector is combined with the secret key to generate the full encryption key Step (4) the RC4 algorithm is used to transform the encryption key into a key stream Step (5) an XOR operation is performed between the key stream and the output of Step (2) Step (6) the initialisation vector is combined with the cipher text. Although WEP provides a reasonable level of security against casual eavesdroppers, its weaknesses were recognised soon after its release as part of the 802.11 standard. In 2001, cryptographers Scott Fluhrer, Itsik Mantin and Adi Shamir realised that, because of a weakness in the RC4 key scheduling algorithm, the output key stream was significantly non-random. This allows the encryption key to be determined by analysing a sufficiently large number of data packets encrypted using the key. In essence, WEP transmits information about the encryption key as part of the encrypted message SO that a determined hacker, equipped with the necessary tools, could collect and analyse transmitted data to extract the encryption key. This requires several million packets to be intercepted and analysed, but could still be accomplished in under an hour on a high traffic network. WEP also uses a static shared key, as there is no mechanism for changing the key other than manual re-entry of a new key or passphrase into every device that operates on the WLAN. Chapter Eight It was not technological limitations that limited the strength of the encryption algorithm used in 802.11 but, interestingly, US export controls. The export of data encrypting technology was considered by the US government to be a threat to national security and as a result the WEP scheme could not be the strongest available if it was to be adopted as an international standard. These restrictions have since been lifted and new, more powerful encryption methods have been developed, such as the

Advanced Encryption Standard (Section "IEEE 802.11i and WPA2, p. 219"). Wi-Fi Protected Access - WPA To overcome these known vulnerabilities in the original 802.11 security implementation, the Wi-Fi Alliance developed Wi-Fi Protected Access (WPA) as a means to provide enhanced protection from targeted attacks. WPA was an interim measure that was based on a subset of the enhanced security mechanisms that were then still under development by 802.11 TGi as part of the 802.11i standard. WPA uses the temporal key integrity protocol (TKIP) for key management, and offers a choice of either the 802. 1x authentication framework together with extensible authentication protocol (EAP) for enterprise WLAN security (Enterprise mode), or simpler pre-shared key (PSK) authentication for the home or small office network which does not have an authentication server (Personal mode). These measures, which were initially available as firmware upgrades to Wi-Fi compliant devices, first came to market in early 2003. In 2004, a further strengthening of the encryption was introduced in the second generation WPA2. This replaced RC4, still used in WPA, with the advanced encryption standard (AES) which was ratified as part of the 802.11i standard in June 2004. The key components of WPA and WPA2 are described in the following sections. Temporal Key Integrity Protocol The WEP encryption vulnerability was addressed in WPA by two new MAC layer features: a key creation and management protocol called TKIP (temporal key integrity protocol) and a message integrity check function (MIC). Features of WEP and WPA key management are compared in Table 8-2. Wireless LAN Security Table 8-2: WEP and WPA Key Management and Encryption Compared Security feature Temporal key/Passphrase 40-bit, 104-bit 128-bit Initialisation vector (IV) 24-bit 48-bit Static Dynamic Encryption cipher After a station has been authenticated, a 128-bit temporal key is created for that session, either by an authentication server or derived from a manual input. TKIP is used to distrib

ute the key to the station and access point and to set up key management for the session. TKIP combines the temporal key with each station's MAC address, plus the TKIP sequence counter, and adds a 48-bit initialisation vector to produce the initial keys for data encryption. With this approach each station will use different keys to encrypt transmitted data. TKIP then manages the update and distribution of these encryption keys across all stations after a configurable key lifetime that might be from once every packet to once every 10,000 packets, depending on security requirements. Although the same RC4 cipher is used to generate an encryption key stream, TKIP's key mixing and distribution method significantly improves WLAN security, replacing the single static key used in WEP with a dynamically changing choice from 280 trillion possible keys. WPA supplements TKIP with a message integrity checking (MIC) that determines whether an attacker has captured, altered and resent data packets. Integrity is checked by the transmitting and receiving stations computing a mathematical function on each data packet. While the simple CRC-32, when used to compute the ICV in WEP, is adequate for error detection during transmission, it is not sufficiently strong to assure message integrity and prevent attacks based on packet forgery. This is because it is relatively easy to modify a message and re-compute the ICV to conceal the changes. In contrast, MIC is a strong cryptographic hash function, which is calculated using source and destination MAC addresses, input data stream, the MIC key and the TKIP sequence counter (TSC). If the MIC value computed by the receiving station does not match the MIC value received in the decrypted data packet, the packet is discarded Chapter Eight Temporal key WEP IV Phase 1 Transmitter key mixing Per packet key MAC address sequence Encrypted counter encapsulation MPDUs MIC key Source address Fragmentation Destination address MSDU plaintext Figure 8-4: TKIP Key Mixing and Encryption Process and counterm

easures are invoked. These countermeasures consist of resetting keys, increasing the rate at which keys are updated, and sending an alert to the network manager. MIC also includes an optional countermeasure, which will deauthenticate all stations and shutdown the BSS for any new association for one minute, if an access point receives a series of altered packets in quick succession. The complete WPA encryption and integrity checking process is shown in Figure 8-4. 802.1x Authentication Framework IEEE 802.1x is an access control protocol that provides protection for networks by authenticating users. After successful authentication, a virtual port is opened on the access point for network access, while communications are blocked if authentication fails. 802.1x authentication defines three elements; The Supplicant - software running on the wireless station that is seeking authentication The Authenticator - the wireless access point that requests authentication on behalf of the supplicant and The Authentication Server - the server, running an authentication protocol such as RADIUS or Kerberos, that provides centralised Wireless LAN Security authentication and access control using an authentication database. The standard defines how the extensible authentication protocol (EAP) is used by the Data Link layer to pass authentication information between the supplicant and the authentication server. The actual authentication process is defined and handled depending on the specific EAP type used, and the access point, acting as an authenticator, is simply a go-between, enabling the supplicant and the authentication server to communicate. Authentication Servers (RADIUS) The application of 802. 1x authentication in an enterprise WLAN, requires the presence of an authentication server within the network, which can authenticate users against a stored list of the names and credentials of authorised users. The most commonly used authentication protocol is the remote authentication dial-in user service (RADIUS), which is supported

by WPA compliant access points and provides centralised authentication, authorisation and accounting services. To authenticate a wireless client seeking network access via an access point, the access point, acting as a client to the RADIUS server, sends a RADIUS message to the server which contains the user's credentials together with information on the requested connection parameters (Figure 8-5). The RADIUS server will either authenticate and authorise or reject the request, in either case sending back a response message. Authenticator Supplicant = Access server Authentication Wireless LAN Intranet / LAN = Client Server RADIUS message format Code Sequence Length Authenticator Attributes Description Access Request Access Accept EAP request and response messages EAP messages encapsulated Access Reject carried as EAPOL packets Accounting Request as RADIUS message attributes Accounting Response Access Challenge Figure 8-5: Message Format for EAP Over RADIUS Authentication Chapter Eight A RADIUS message comprises a RADIUS header and RADIUS attributes, with each attribute specifying a piece of information about the requested connection. For example, an Access-Request message will contain attributes for the user name and credentials, and the type of service and connection parameters being requested by the user, while the Access-Accept message contains attributes for the type of connection that has been authorised, relevant connection constraints and any vendor specific attributes. Extensible Authentication Protocol The extensible authentication protocol (EAP) builds on the framework for enabling remote access that was originally established for dial-up connections in the point-to-point protocol (PPP) suite of protocols. The PPP dial-up sequence provided for the negotiation of link and network control protocols, as well as the authentication protocol that would be used, based on the desired level of security. For example, an authentication protocol, such as password authentication protocol (PAP) or challenge- handshake

authentication protocol (CHAP), is negotiated between client and the remote access server when a connection is established and then the chosen protocol is used to authenticate the connection. EAP extended this structure by allowing the use of arbitrary authentication mechanisms, called EAP types, which define various structures for the authentication message exchange. When a WLAN connection is being established, client and access point agree on the use of EAP for authentication, and a specific EAP type is chosen at the start of the connection authentication phase. The authentication process then consists of the exchange of a series of messages between the client and authentication server, the length and detail of the exchange depending on the requested connection parameters and the selected EAP type. Some of the most common EAP types are described below. When EAP is used together with RADIUS as the authentication protocol, EAP messages sent between the access point and the authentication server will be encapsulated as RADIUS messages, as shown in Figure 8-5. Extensible Authentication Protocol over LANs To apply EAP to LANs or WLANs rather than to dial-up connections, extensible authentication protocol over LAN (EAPoL) was defined in the Wireless LAN Security 802.1x standard as a transport protocol for delivering authentication messages. EAPoL defines a set of packet types that carry authentication messages, the most common of which are; EAPoL-Start - Sent by the authenticator to start an authentication message exchange EAP-Packet - Carries each EAP message EAPoL-Key - Carries information related to generating keys EAPoL-Logoff - Informs the authenticator that the client is logging off. EAP Types EAP types supported by the Wi-Fi Alliance's interoperability certification programme include; EAP-TLS, EAP-TTLS/MS-CHAP v2, PEAP v0/EAP-MS-CHAP v2, PEAP v1/EAP-GTC and EAP-SIM. To give a flavour of how these EAP types differ, EAP_TLS, EAP-TTLS and PEAP are briefly described here. EAP-TLS (Transport layer security) uses c

ertificate based authentication between client and server, and can also dynamically generate keys to encrypt subsequent data transmissions. An EAP-TLS authentication exchange requires both the station and the authentication (RADIUS) server to prove their identities to each other using public key cryptography and the exchange of digital certificates (see next section). The client station validates the authentication server's certificate and sends an EAP response message that contains its certificate and starts the process of negotiating encryption parameters, such as the cipher type that will be used for encryption. As shown in Figure 8-6, once the authentication server validates the client's certificate, it responds with the encryption keys to be used during the session. EAP-TLS therefore requires initial configuration of certificates on both the client station and the authentication server, but once this is established by the network manager no further user intervention is required. On the client station, the certificate must be protected by a passphrase or PIN, or stored on a smart card. The result is a very high level of wireless security although, for large WLAN installations, the requirement to manage both client and server certificates as well as the Wireless LAN Intranet / LAN Supplicant Authenticator Authentication = Client = Access server Server EAPOL Start EAP Identity Request Request Client for Identity Authenticator makes Access EAP Identity Response RADIUS Access Request Request with User ID EAP - TLS Start Start EAP - TLS exchange EAP Response (Client -Hello) EAP Request (Request client certificate) Server sends certificate EAP Response (Client certificate and cipher specification) Client verifies Server certificate EAP Request (Cipher specification) Server verifies Client certificate EAP Response TLS finished EAP Success RADIUS Access Success Client derives Session key EAPOL Multicast Key Authenticator delivers multicast EAPOL Session Parameters key encrypted with session key Figure 8-6: EAP-TLS Au

thentication Exchange Wireless LAN Security public key infrastructure (PKI) required for certificate validation can become a significant network management task. EAP-TTLS (Tunnelled transport layer security) and PEAP (Protected extensible authentication protocol) are alternatives to EAP-TLS that dispense with the client side certificate, and the associated implementation and administration overhead. The client station confirms the identity of the authentication server by verifying its digital certificate using PKI, and then a one-way TLS tunnel is set-up allowing the client's authentication data (password, PIN, etc.) to be encapsulated as TLS messages and securely transported to the authentication server. Public Key Infrastructure A public key infrastructure (PKI) enables digital certificates to be used to electronically identify an individual or an organisation. A PKI requires a certificate authority (CA) that issues and verifies digital certificates, a registration authority (RA) that acts as the verifier of the CA when a new digital certificate is issued, and a certificate management system, including one or more directory services where the certificates and their public keys are stored. When a certificate is requested, a CA simultaneously creates a public and private key using an algorithm such as RSA (see Glossary). The private key is given to the requesting party and the public key is lodged with a publicly available directory service. The private key is held securely by the requesting party and is never shared or sent across the Internet. It is used to decrypt messages that have been encrypted with the related public key, accessed from the public directory by the party sending the message. A PKI enables a user to digitally sign a message by encrypting using a private key, and allows a recipient to check the signature by retrieving the sender's public key and using this to decrypt the message. In this way the parties can establish user authentication, as well as message privacy and integrity, without the ne

ed to exchange a shared secret. IEEE 802.11i and WPA2 The IEEE 802.1 11i standard defines security enhancements for 802.11 WLANs, providing stronger encryption, authentication and key management Chapter Eight strategies with the objective of creating a robust security network (RSN). The key features of the RSN are; A negotiation process that enables the appropriate confidentiality protocol for each traffic type to be selected during device association A key system that generates and manages two hierarchies of keys. Pairwise keys for unicast and group keys for multicast messages are established and authenticated through EAP handshakes during device association and authentication Two protocols to improve data confidentiality (TKIP and AES-CCMP). Key caching and pre-authentication are also included in 802.11i to reduce the time taken for roaming wireless stations to associate or re-associate with access points. WPA2 is the Wi-Fi Alliance's implementation of the final IEEE 802.11i standard and replaced WPA following the ratification of 802.11i in June 2004. WPA2 implements the advanced encryption standard (AES) encryption algorithm using counter mode with cipher block chaining message authentication code protocol (CCMP). TKIP and 802.11 authentication were included in the earlier release of WPA, and have been described above, while AES and CCMP are covered in the following sections. WPA and WPA2 both support Enterprise and Personal modes, and a comparison of the main elements is shown in Table 8-3. Table 8-3: WPA and WPA2 Compared Enterprise mode Personal mode Authentication: IEEE 802.1x/EAP Authentication: PSK Encryption: TKIP Encryption: TKIP Integrity: MIC Integrity: MIC Authentication: IEEE 802.1x/EAP Authentication: PSK Encryption: AES-counter mode Encryption: AES-counter mode Integrity: CBC-MAC (CCMP) Integrity: CBC-MAC (CCMP) Wireless LAN Security RSN Security Parameter Negotiation The negotiation of security parameters between RSN capable devices is enabled by including an RSN information element (IE) which i

dentifies the broadcasting device's RSN capabilities in beacon, probe, association and reassociation frames. The IE identifies specific RSN capabilities; supported authentication and key management mechanisms, and ciphers for unicast or multicast messages, as described in Table 8-4. The selection of security parameters occurs through the following exchange; Step (1) The client station broadcasts a Probe Request. Step (2) The access point broadcasts a Probe Response including an RSN IE. Step (3) The client station sends an Open System Authentication request to the access point. Step (4) The access point provides an Open System Authentication response to the client. Step (5) The client station sends an Association Request with an RSN IE indicating its choice of RSN capabilities. Step (6) The access point send an Association Response indicating success if the client station's selected security parameters are supported by the access point. Notice that this exchange is unprotected and, since Open System Authentication is used (see the Section "Station Services, p. 147"), there is effectively no authentication between the client station and the access Table 8-4: RSN Information Element content RSN capability Description Supported authentication and RSN devices can either support 802.1x key management authentication and key management or mechanisms 802. 1x key management with no authentication Supported ciphers RSN devices may support any of the following ciphers for either unicast or multicast message encryption (WEP, TKIP, WRAP and AES- CCMP) Chapter Eight point. This does not introduce a security threat since the exchange simply serves to establish the protocol and cipher that will then be used to ensure mutual authentication and subsequent data privacy. To provide a degree of backward compatibility with legacy equipment, networks that use RSN but do not have the required hardware to support AES will allow the use of TKIP/RC4 for encryption. This interim step towards full RSN is referred to by the term transition sec

urity network (TSN). RSN Key Management After security parameter negotiation, the next stage in establishing a connection between a client station and an access point is mutual authentication using 802. 1x or PSK authentication, as described in the Section "802.1x Authentication Framework, p. 214". At the end of this authentication exchange, the authentication server generates a pairwise master key or alternatively, in personal mode, the key is derived from the user entered password or passphrase. Pairwise Key Hierarchy and the 4-way Handshake Pairwise keys are used to protect unicast messages between a client station and access point. The hierarchy of keys and the handshake that establishes and installs them are described in Table 8-5. The purpose of the 4-way handshake, shown in Figure 8-7, is to install this key hierarchy securely in both the client station (supplicant) and the access point (authenticator). Each of the four steps involves the transmission of an EAPoL key exchange message, as follows; Step (1) The access point (authenticator) generates a pseudo-random nonce (ANonce) and sends this to the supplicant. Step (2) The client station (supplicant) generates a pseudo-random nonce (SNonce) and is then able to compute the PTK and derive the KCK and KEK. The supplicant sends its SNonce to the authenticator, together with the security parameters previously negotiated. The KCK is used to compute a MIC that assures the origin of the message. Wireless LAN Security Table 8-5: Pairwise Key Hierarchy Pairwise keys Description Pairwise master key (PMK) Starting point of the pairwise key hierarchy, either generated by the authentication server or derived from the user entered password if 802. 1x authentication is not being used. Pairwise transient key (PTK) The PTK is derived from the PMK together with the MAC addresses of the client station and access point, and a nonce (see Glossary) provided by each party during the 4-way handshake. EAPoL key confirmation key The KCK is used to assure authenticity of messages (K

prevent it from receiving any further broadcast or multicast messages from the access point. The current GTK is shared with an associating client station in the third EAPoL exchange of the 4-way handshake, but a further handshake, the group key handshake, is used when the GTK needs to be updated. A new GTK will be derived by the access point using a pseudo-random function of the GMK together with its MAC address and a nonce (GNonce). The new GTK is then distributed via the group key handshake, as follows; Step (1) The access point sends the new GTK in encrypted unicast messages to each station in the BSS. The new GTK is encrypted using each station's unique KEK and protects the data from being tampered using a MIC. Wireless LAN Security Step (2) Each station replies to inform the access point that the new GTK is installed. All stations will then use the new GTK to decrypt future broadcast or multicast messages. Advanced Encryption Standard (AES) The Advanced Encryption Standard (AES) is a cipher that was developed by Belgian cryptographers Joan Daemen and Vincent Rijmen, and was adopted as an encryption standard in November 2001 by the US National Institute of Standards and Technology (NIST) after a four year selection process. In June 2003, the US Government authorised the use of AES to protect classified information, including "Top Secret" information, provided that either 192 or 256 bit key lengths were used. Unlike RC4, which is a stream cipher and can encrypt a message of arbitrary length, AES is a block cipher and uses a fixed message block size of 128 bits together with an encryption key of 128, 192 or 256 bits. This is a specific instance of Daemen and Rijmen's original cipher, also known as the Rijndael cipher, which can use block and key sizes of 128 to 256 bits, in steps of 32 bits. The cipher operates on 4 X 4 arrays of bytes (i.e. 128-bits), and each round of the cipher consists of four steps; Step (1) A SubByte step - where each byte in the array is substituted with its entry in a fixed 8-bit lookup

table called the S-box. A ShiftRows step - where each row of the 4 x 4 array is shifted a certain number of positions, with the size of the shift differing for each row of the array. Step (3) A MixColumns step - which mixes the four bytes in each column of the array using a linear transformation to produce a new column of four dependent output bytes. This step is omitted in the final round of the cipher. Step (4) An AddRoundKey step - in which a second 4 X 4 array, called a sub key, is derived from the cipher key using a key schedule and the two 4 X 4 arrays are XOR'd together to generate the starting array for the next round. Chapter Eight The number of rounds used to encrypt each block of data depends on the key size, with 10 rounds used for 128-bit, 12 rounds for 192-bit and 14 rounds for 256-bit keys. While AES is already a very strong cipher, AES-CCMP incorporates two additional cryptographic techniques, counter mode and a cipher block chaining message authentication code (CBC-MAC) that provide additional security between the wireless client station and the access point. Counter Mode Operation of Block Ciphers In the counter mode of operation of a block cipher, the encryption algorithm is not applied directly to a block of data but to an arbitrary counter. Each block of data is then encrypted in an XOR operation with the encrypted counter, as shown in Figure 8-8. For each message, the counter is started from an arbitrary nonce and incremented according to a pattern that is known to both sender and receiver. Counter mode contrasts with the electronic code book (ECB) (Figure 8-9) mode of operation that was commonly used with the data encryption standard (DES - the NIST predecessor of AES) in which each block of a message is encrypted with the same encryption key. With ECB, the output ciphertext has a one-to-one correspondence with the input plaintext Message Data block Data block Data block Data block Data block Data block Data block Counter Encryption Encrypted Encrypted Encrypted Encrypted Encrypted Encrypt

ed Encrypted Ciphertext data block data block data block data block data block data block data block Figure 8-8: Counter Mode of Operation of a Block Cipher (AES) Wireless LAN Security Message Data block Data block Data block Data block Data block Data block Data block Encryption Encrypted Encrypted Encrypted Encrypted Encrypted Encrypted Encrypted Ciphertext data block data block data block data block data block data block data block Figure 8-9: Electronic Code Book (ECB) Mode of Operation of a Block Cipher (AES) message, SO that patterns in the input data are preserved as patterns in the encrypted data, simplifying the task of discovering the encryption key. Counter mode removes this correspondence and eliminates the risk of key discovery as a result of such patterns. Wireless robust authenticated protocol (WRAP) is a further security mechanism supported by RSN, and is based on the so-called offset codebook (OCB) mode of operation of a block cipher. This scheme has a number of advantages over AES-CCMP, including being computationally more efficient, as it performs message integrity checking, authentication and encryption in a single calculation. However, WRAP was not adopted by TGi as the base security protocol for RSN because OCB is a patented scheme, which would have introduced licensing complications into the standard. Cipher Block Chaining Message Authentication Code (CBC-MAC) CBC-MAC is a message authentication and integrity method that can be used with block ciphers such as AES. The acronym MIC will be used instead of MAC for the Message Authentication Code, to avoid confusion with MAC as in Media Access Control. Cipher block chaining (the CBC part) is a mode of operation of a block cipher in which the ciphertext of one block becomes part of the encryption algorithm for the next block, and SO on. The CBC-MAC message authentication code is generated as shown in Figure 8-10, with the following steps; Step (1) A 104-bit nonce is created by combining an 8-bit priority field with the 48-bit source MAC address

and a 48-bit packet number. Chapter Eight 8-bit Priority 48-bit MAC address 48-bit Packet number 104-bit Nonce 8-bit 104-bit Nonce 16-bit Dlen 128-bit starting block 128-bit Starting block 128-bit Data block 128-bit Data block 128-bit Data block counter mode encryption Result 1 counter mode counter mode encryption encryption Result 2 Result 3 64-bit MIC 64-bit discard Figure 8-10: MIC Computation Using AES and CBC-MAC Step (2) The nonce is concatenated with an 8-bit flag and a 16-bit Dlen field, which indicates the unpadded length of the plaintext data field, to generate a 128-bit starting block for block chaining, Step (3) The starting block is encrypted using AES in counter mode (Figure 8-8) and the result is XOR'd with the first block of plaintext data from the message, Step (4) The result is encrypted using AES in counter mode and the result is XOR'd with the second block of plaintext data, Step (5) This chaining is repeated until the last plaintext block is XOR'd, Step (6) The higher 64-bits of the 128-bit result are the MIC and the lower 64-bits are discarded. Robust Security Network and AES-CCMP The RSN security protocol, AES-CCMP (or AES counter mode-CBC- MAC protocol) defines how the three elements, AES, counter mode and CBC-MAC are used to protect data in an 802.11i implementation. The encryption of a MAC protocol data unit (MPDU), consisting of a MAC header and a data packet, proceeds as follows (Figure 8-11); Wireless LAN Security Packet number CCMP header GTK ID MAC header CCMP header CBC-MAC Counter mode encryption MAC header CCMP header Ciphertext Figure 8-11: MPDU Encryption Using CCMP Step (1) A CCMP header is constructed from a 48-bit packet number and a 3-bit GTK ID, and inserted into the MPDU, between the MAC header and the data payload. Step (2) A CBC-MAC is computed on this extended MPDU, as described in the previous section, and the resulting MIC is appended to the plaintext data block. Step (3) The resulting, so-called encoded data block (data + MIC) is encrypted using AES in counter mode.

Step (4) The ciphertext is appended to the unencrypted MAC and CCMP headers. Although the MAC and CCMP headers are transmitted in plaintext, the MIC protects both the data packet and the headers from error or malicious alteration (Step (2) above). The Sections "WEP - Wired Equivalent Privacy Encryption, p. 209", "Wi-Fi Protected Access; WPA, p. 212" and "IEEE 802. 11i and WPA2, p. 219" have described the key technologies that have been developed to enable wireless LANs to achieve a high level of security. However, that security will only be assured if these technologies are supported in turn by the necessary practical measures. The following section turns to these practical security measures. Chapter Eight WLAN Security Measures In order to ensure that all security vulnerabilities are recognised and addressed, every WLAN implementation needs to consider three aspects of security - management, technical and operational. A comprehensive checklist of best practice security measures in these three areas has been published by the US National Institute of Science and Technology (NIST Special Publication 800-48, see Resources the Section "General Information Sources, p. 363"). Management Security Measures Management security measures address issues that need to be considered when designing and implementing a WLAN. Some of the key management security measures recommended as best practice in the NIST checklist are described in Table 8-7. Table 8-7: WLAN Management Security Measures Management security measure Description Develop a security policy for the The security policy provides the foundation for organisation that addresses a secure WLAN and should specify the the use of wireless technology organisation's requirements including access control, password usage, encryption, control of equipment installation and administration, etc. Perform a risk assessment to Understanding the value and the potential understand the value of the consequences of unauthorised access to assets in the organisation the organisation's assets

will provide the that need protection basis for establishing the required level of security. Take a complete inventory of all A physical inventory of installed devices should access points and wireless be cross-checked with WLAN logs as well as devices periodic RF sweeps for unknown devices (rogue access points). Locate access points on the interior Internal location will limit the leakage of RF of buildings instead of near transmissions beyond the required operating external walls and windows area and eliminate areas where eavesdropping could take place. Place access points in secured areas Physical security will prevent unauthorised access and manipulation of hardware. Wireless LAN Security Technical Security Measures Technical security measures address issues that need to be considered when configuring a WLAN. Table 8-8 describes some of the key technical security measures recommended in the NIST checklist. Changing SSID and Disabling Broadcasts Unless disabled, the SSID is included in beacon frames transmitted by an access point about ten times every second to alert nearby stations of its presence. Every access point leaves the factory with a default SSID set, and attackers can use these IDs to access unsecured networks if they are not changed from the default values. These default values are widely known and published on the web, and a few of the more well-known ones are "tsunami", "linksys", "wireless" and "default"! War Drivers and War Flyers report that 60-80% of identified WLANs typically still have their security setting in factory default mode, 60% have not changed default SSIDs and under 25% have enabled WEP. These figures are reported from the US, but would likely be similar in the UK. Changing the SSID to another value is the first step in improving security. This should be done when the access point is first configured (Figure 8-12), since entering the SSID is part of the configuration process for each of the client stations that will connect to the access point. A good practice is to use an anony

mous SSID that does not identify the company or organisation operating the WLAN. After changing the default SSID, some manufacturers provide an option to disable the SSID from being broadcast in beacon frames. This will prevent the casual eavesdropper from obtaining the SSID, but will not stop the determined hacker. The SSID is included unencrypted in the Probe Response frame transmitted by an access point when responding to a Probe Request sent by a client station attempting to connect, SO a hacker equipped with "sniffer" software can extract the SSID from these messages. The SSID is not intended to perform a security function, and changing the SSID from its default and disabling SSID broadcasts will only be effective against casual unauthorised access. Stronger measures are required to guard against a targeted attack. Chapter Eight Table 8-8: WLAN Technical Security Measures Technical security measure Description Change the default SSID and disable Prevents casual access to the WLAN and SSID broadcast requires a client station to match the SSID when attempting to associate. Disable all nonessential management Each management protocol provides a protocols on access points possible route of attack, SO disabling unused protocols minimises the potential routes that an attacker could use. Ensure that default shared keys are Manual key management will be necessary replaced by keys of at least unless TKIP is installed (Section "Temporal 128-bits and periodically Key Integrity Protocol, p. 212"). Best practice change keys is to use the longest supported key length. Deploy MAC access control lists Access control based on MAC filtering provides additional security although, as described in the Section "MAC Addressing Filtering, p. 234", it is not secure against a technically determined attacker. Enable user authentication and strong Management control functions on access administrative passwords for access points need to be protected as well as, if not point management interfaces better than, the network traffic. The sec

urity policy should specify the requirement for user authentication and strong passwords. Changing Default Shared Keys Although the fundamental cryptographic weakness has been recognised and addressed, as was described in the Section "WEP - Wired Equivalent Privacy Encryption, p. 209", it is still good security practice to enable data encryption using wired equivalent privacy (WEP) in pre-802.11i networks, and to use the longest supported shared key length. Some access points have factory set default encryption keys and, similar to SSIDs, it is recommended practice to change these default values when encryption is enabled. Figure 8-13 shows a typical access point configuration screen, with encryption keys being generated using an algorithm based on a password or passphrase entered on set-up. Shared keys or passphrases must be changed to maintain security whenever a previous network user is no longer authorised to access the network or if a device that has been configured to use the network is lost or compromised. Wireless LAN Security D-Link Building Networks for People Air Plus G 802.11g/2.4GHz Wireless Router Advanced Tools Status DI-524 Wireless Settings These are the wireless settings for the AP(Access Point) portion. Network ID(SSID) Another_SSC Channel Wizard Security Wireless Apply Cancel Help Figure 8-12: Changing SSID Default D-Link Bunding Networks for People AirPlus G 802.11g/2.4GHz Wireless Router Advanced Tools Status DI-524 Wireless Settings These are the wireless settings for the AP(Access Point) portion Network ID(SSID) Another _SSID Channel Wizard Security WEP Encryption 64 Bit Wireless Key Mode WEP Key A20B30CC40 Input 10 HEX characters (HEX is 0~9, A~F, or a~D Apply Cancel Help Figure 8-13: Enabling WEP and Specifying Shared Keys Chapter Eight MAC Address Filtering If MAC address filtering is enabled, the access point will review every request for access against a permit list held in its memory. The permit list holds the MAC addresses of all authorised client stations, and the access point will

only grant access if the requesting MAC address is found in the list. MAC filtering appears to be a highly effective security measure but, as MAC addresses are included in transmitted data packets, a hacker could recover permitted MAC addresses by "sniffing" wireless traffic. Although the MAC address is a factory set identifier, that is, in principle, unique to an individual adapter card, a MAC address can also be temporarily reset by software to allow a device to masquerade as (or spoof) another device. This may sound like a built-in fault, but it also has a legitimate use as some ISPs use MAC address filtering to control customer access. If it is necessary to connect more than one device to the Internet using a single ISP connection it will be necessary to reset the MAC addresses of all devices to be the same as the MAC address that was registered with the ISP. Enabling MAC address filtering is a simple button click in the access point set-up procedure, as shown in Figure 8-14. However, keeping the access list up-to-date can be a time consuming administration task, as every authorised client station must have its MAC address manually entered into the list. In a home WLAN this will be easy enough to maintain, but in a dynamic network such as an enterprise or community WLAN, where clients may be changing fairly frequently, the cost of the administration burden may quickly outweigh the security benefits it provides. Operational Security Measures Operational security measures address issues that should be considered during routine operation of a WLAN. The key operational security measures recommended in the NIST checklist are summarised in Table 8-9. Strong User Authentication Remote authentication dial-in user service (RADIUS), as described in the Section "802.1x Authentication Framework, p. 214", is the most Wireless LAN Security D-Link Building Networks for People AirPlus G 802.11g/2.4GHz Wireless Router Advanced Tools Status DI-524 Filter Filters are used to allow or deny LAN users from accessing the Internet

IP Filters URL Blocking MAC Filters Domain Blocking Virtual Server MAC Filters Use MAC address to allow or deny computers access to the network Application Disabled MAC Filters Filter Only allow computers with MAC address listed below to access the network Only deny computers with MAC address listed below to access the network Firewall Blocked computer 2 MAC Address DHCP Client select one Clone Apply Cancel Help Performance MAC Filter List MAC Address Blocked computer 1 00-40-05-7A-AA-40 Figure 8-14: Enabling MAC Address Filtering commonly used authentication protocol, providing centralised authentication, authorisation and accounting services. Kerberos is another authentication protocol, developed by the Massachusetts Institute of Technology, and provides tools for authentication and strong encryption particularly for client/server applications. The source code is freely available from MIT and has also been incorporated into a range of commercial products. Intrusion Detection Intrusion detection software can be run to continuously monitor activity on the WLAN and generate an alarm when any unauthorised device, such as a rogue access point, is detected. Typically intrusion monitoring is based on specifying parameters that identify normal, authorised WLAN devices and traffic, as shown in Table 8-10. When devices or network activity is identified that deviates from the specified parameters, an alarm is generated that can either be used to Chapter Eight Table 8-9: WLAN Operational Security Measures Operational security measure Description Use an encrypted protocol, such as SNMP v3 provides encryption of access point SNMP v3, for access point management messages, whereas SNMP v1 configuration and v2 do not provide the same level of security. Consider other forms of user If a risk assessment identifies unauthorised authentication for the wireless access as a key risk, authentication services or network such as RADIUS and protocols, such as RADIUS and Kerberos, can Kerberos provide a high degree of access security to p

rotect confidential data. Deploy intruder detection on Rogue access points or other unauthorised the WLAN to detect activity can be detected by intrusion detection unauthorised access or software. This is a standard feature of activity wireless switches, described in the Section "Wireless LAN Switches or Controllers, p. 48". When hardware is upgraded If access points are left with their secure ensure that configuration settings configuration settings when they are disposed are reset prior to disposal of, this sensitive information could be used of old equipment to attack the network. Enable and regularly review Access point logs provide a basis for periodic access point logs auditing of network traffic - both authorised and unauthorised. Many intrusion detection tools can be configured to effectively perform this task automatically. refine the parameter list or to warn the WLAN manager of an intrusion attempt and initiate countermeasures such as blocking an identified rogue device from associating or maintaining a connection with the WLAN. Intrusion detection software can also monitor network activity to detect any attacks on the WLAN, such as DoS attacks or session highjacking, and to ensure that all authorised devices are complying with the security policies in force. Wireless Hotspot Security Wi-Fi hotspots provide public wireless connections to the Internet at convenient locations such as coffee shops, hotel and airport lounges. The requirement for easy public accessibility means that encryption is not Wireless LAN Security Table 8-10: Intrusion Detection Parameters Security parameter Description Authorised RF PHY specifications Identifies which PHY layer standards are being operated in the WLAN (e.g. 802.11a, 802.11b/g). Authorised RF channel usage Indicates which RF channels individual access points are configured to use. Authorised device MAC addresses Similar to MAC access control at the access point level, but extended to the whole WLAN. SSID policy Lists authorised SSIDs. Equipment vendor Indicates the

manufacturer of the equipment that is authorised to operate on the WLAN. This provides partial MAC filtering, since the first part of the MAC address indicates the equipment manufacturer. enabled since, prior to 802.11i, the 802.11 standards lacked the necessary key management mechanisms. As a result, responsibility for the security of information sent over the wireless connection rested firmly with the hotspot user. With the ratification of the 802.11i standard in 2004, the next generation of hotspot services will have the full range of security measures available, including the option of EAP and RADIUS based authentication and the full 802.11i pairwise and group key management mechanisms described in the Section "Wi-Fi Protected Access; WPA, p. 212". Until secure hotspot services become available, the technical and operational security measures described in Table 8-11 should be considered when using public hotspots. Secure Socket Layer Secure socket layer (SSL) is a layer 6 protocol that was developed by Netscape to protect confidential data being transmitted over the Internet, and relies on the public key infrastructure (PKI) to establish encryption keys. SSL supports several public key algorithms, including RSA (see Glossary), and several encryption ciphers, including RC4 and AES, the choice being negotiated when the secure connection is set up. Chapter Eight Table 8-11: Wireless Hotspot Security Measures Hotspot Security Measure Description Set up the wireless network Limiting automatic connection to an identified connection to only connect list of preferred access points will reduce the to preferred access points risk of connecting to an unknown access point. However, as SSIDs can be easily spoofed, this will not eliminate the risk from rogue access points. Use VPN for connecting to A VPN uses an additional level of encryption to a corporate network provide a protected "tunnel" through an insecure connection such as the Internet or a wireless hotspot connection. Install a personal firewall on A firewall act

s as a barrier to prevent unauthorised mobile PCs that use hotspots access to a mobile computer through the hotspot connection. Information received from the access point will be permitted or blocked depending on configuration of the firewall. Wireless network traffic should be assigned an "untrusted" status when using a public access point. Protect files and folders on the Using the privacy mechanisms available in the mobile device, using PC operating system will ensure that files passwords or encryption. and data are protected even if an attacker Disable file sharing. manages to make an unauthorised connection to the mobile device. Protect data transmitted to the Confidential data, including e-mails, should be access point encrypted before transmission, or a secure socket layer (SSL) e-mail service can be used. Disable the wireless NIC when Turning off the wireless NIC radio when not in not in use use will remove a potential attack route and conserve battery power for mobile devices. Beware of the surveillance risk Be vigilant when entering PINs or passwords in in public places public places to ensure that confidential information is protected from surveillance. Keep operating system and Security patches should be downloaded regularly security software up-to-date to ensure that all software contributing to security - operating system, firewall and anti-virus software - is updated to combat all known threats. Wireless LAN Security Web sites and secure on-line services such as e-mail that use SSL security can be identified by the https:// URL prefix and by a padlock icon displayed in the browser window. The digital certificate being used by a secure web site or service provider can be checked by double clicking the padlock icon, which will display certificate information such as the name and e-mail address of the certificate owner, certificate usage and validity date, and the web site address or e-mail address of the resource using the certificate. The certificate ID of the person or entity that certifies the inf

ormation contained in the first certificate will also be provided, SO that the full certification path can be checked. VoWLAN and VolP Security As noted in the previous chapter, when telephony moves from the relative security of physical telephone lines and exchanges to the wireless LAN or Table 8-12: VoWLAN/VolP Security Measures Vo WLAN/VOIP security measure Description Separate VoWLAN phones Putting VoWLAN phones on a separate virtual from other LAN devices LAN (VLAN) and using private, non-routable, using a virtual LAN IP addresses will prevent the VoWLAN system from being accessed or attacked via the Internet. IP phones will be needed that support the 802.1p/q VLAN standard. Encrypt voice traffic between Encryption can eliminate voice traffic's the phone and the PSTN vulnerability to eavesdropping en route to the gateway telephone network. Use a firewall to protect If a connection between voice and data WLANs connections between voice is required, for example, to allow desktop and data WLANs software to manage VoWLAN phone information, use a properly configured firewall to eliminate unwanted access routes (ports, protocols, etc.). Use a VPN to secure VoIP A separate user log-in to the VoWLAN virtual traffic from remote users LAN can be used to ensure that access is limited over the Internet to only the necessary ports and protocols. Keep VoWLAN phone Ensure that there is a management system in software up-to-date place to keep phone software up-to-date with security patches. Chapter Eight the unsecured Internet, additional security measures are required to secure confidential voice traffic. Specific measures that may be taken to secure voice traffic are summarised in Table 8-12, and should be addressed in the WLAN security policy before voice services are introduced. Summary From the relatively simple beginnings of 802.11 security based on WEP and the quickly discovered cryptographic weaknesses of the underlying key scheduling algorithm, the pace of development and current level of sophistication of WLAN sec

urity has paralleled the advances in technology aimed at increasing speed, network capacity and other functional capabilities. With the ratification of the 802.11i security enhancements in 2004, and the launch of the WPA2 implementation by the Wi-Fi Alliance, the foundations are in place for future WLAN installations to deliver true wired equivalent privacy over the inherently open wireless medium. However, security is only assured when these powerful technological capabilities are supported by properly implementing the practical security measures described in this chapter. CHAPTER Wireless LAN Troubleshooting Analysing Wireless LAN Problems As with any problem-solving exercise, resolving wireless LAN problems requires a systematic approach, first in analysing the symptoms to try to narrow down the possible cause, and then in investigating potential solutions. If a step-by-step approach to network implementation was followed, as described in Chapter 7, then troubleshooting will typically be called for when a change in network or device performance occurs - either something worked before and now does not, or now does not perform as it did before. Start by asking a number of questions to clarify the nature and extent of the problem (Table 9-1). This will help to narrow down the range of possible root causes. The starting point for diagnosis should be a review of any recent changes in the network hardware or configuration, in the operating environment or pattern of usage, as summarised in Table 9-2. Answering some of these questions may require further investigation, such as a repeat RF site survey if increased RF interference is a possible cause of performance degradation. The systematic approach should continue when testing possible solutions to the problem, using the strategies described in Table 9-3. The majority of WLAN problems fall into two categories, connectivity - when one or more client stations are unable to establish a connection to the network, and performance - when the data throughput and response ti

me of the network does not match user expectations or previous Chapter Nine Table 9-1: WLAN Troubleshooting - Narrowing down the Problem Problem identification Considerations Is it a connection problem? A major category of problems with WLANs relate to client station connections; individual users or groups of users may be unable to connect to previously accessible network resources. Is it a performance problem? The second major category of problems is perform- ance related; network coverage, speed or response time are not as expected or as previously experienced. How extensive is the problem? Does the problem affect just one device or is the same problem experienced by many? For example, if the problem is connectivity to the network, if only one client is affected, suspect the hardware and set-up of that device's NIC. If a whole BSS is affected, check the hardware and configuration of the access point. How regular is the problem? Does it occur continuously, at specific times of the day - for example around mid-day when staff are using a microwave oven in the lunch room? experience. These two categories of problems are considered in the following sections. Connectivity Problem Checklist Connectivity problems can have a root cause at the PHY or MAC level, for example due to physical or configuration problems with RF hardware, or at higher levels, for example due to a failure during the user authentication process. The checklist shown in Table 9-4 can be used as a starting point for diagnosing connectivity problems. Performance Problem Checklist Performance problems occur in WLANs either because a transmission is not reaching the receiving station with sufficient signal to noise ratio (SNR) to be properly detected and decoded, or because an access point is overloaded and unable to cope with the volume of traffic. In turn, SNR problems can be due to low signal (coverage holes) or high noise (interference). The checklist in Table 9-5 can be used as a starting point to address performance problems. Wireless LAN Trouble

shooting Table 9-2: WLAN Troubleshooting - First Considerations Recent WLAN changes Considerations Hardware changes Have any new hardware devices been added to the network? Is the new hardware from the same manufacturer as existing hardware, or one that has been certified for interoperability? Configuration changes Have any configuration settings been changed recently? Operating channel changes, security mechanisms enabled, keys or passphrases changed? Software changes Have software or firmware upgrades recently been installed? Are any set-up changes required as a result of newly installed patches to client computer or network operating systems, device drivers or firmware? Environmental changes - Has any hardware recently been moved, such as an physical access point possibly creating an RF coverage hole? Have partitions walls or furniture (metal filing cabinets) been rearranged in the operating area, potentially affecting the RF propagation pattern? Environmental changes - Have any new wireless networks or other RF sources RF environment been installed in the operating environment or in the neighbourhood? (For example a fast-food restaurant in the next building with microwave ovens along the shared wall.) Usage pattern changes Have any new applications been installed which make use of the WLAN, particularly those requiring high continuous or peak bandwidth? Have there been any changes in network usage, for example a new user group with high bandwidth demands? Troubleshooting using WLAN Analysers Dedicated WLAN analysers are available to monitor and troubleshoot enterprise scale installations. These systems include tools for site surveying, security assessment, network performance monitoring and troubleshooting that can help the network manager in the tasks of designing, implementing, securing and finally troubleshooting WLANs. Analysers are available either as stand-alone hardware or as software packages that can be run on a laptop or handheld computer. Some examples are shown in Figure 9-1. Chapter Nine Table 9-

3: WLAN Troubleshooting - Solution Strategies Solution approach Description Test one hypothesis at a time Whether it is a configuration setting or a physical layout that is being tested, make changes one at a time SO that effects can be directly attributed to a single cause. Test hardware using a known The easiest way to identify a hardware fault is to functioning substitute replace the suspected item with a known functioning substitute - whether it is a length of CAT 5 cable, a NIC or an access point. Keep a record Make a note of the changes that are made, any initial settings that are altered and the resulting response of the system. This will ensure that time will not be wasted in reinvestigating old avenues and that the previous set-up can be reinstated if changes make things worse. Check for unexpected side Before declaring a problem solved, check as far effects as possible that new unwanted symptoms have not been introduced as a consequence of the solution of the original problem. When everything else fails Read or re-read the hardware vendor's installation read the instructions instructions and check their Web site for specific information on problem diagnosis and troubleshooting. Figure 9-1: WLAN Network Analysis Tools (Courtesy of Fluke Networks) Wireless LAN Troubleshooting Table 9-4: WLAN Troubleshooting - Checklist for Connectivity Problems Problem symptoms Checkpoints A single user is unable Check that the wireless NIC is not disabled and that the to connect to any access station has adequate received signal strength point Check whether another client station is able to connect at the problem location Check the configuration of the client station's wireless network connection, including security settings Check that access point security mechanisms such as MAC filtering are correctly configured for the client station Replace any suspect wireless NIC with a known functioning substitute No user is able to connect Check the configuration of the access point, including to an access point security settings

Recheck connectivity with security settings temporarily disabled Replace the suspect access point with a known functioning substitute Users can connect to an Check that the client station and access point have valid access point but are IP addresses, sub-net masks and default gateway unable to access the addresses, either received from a DHCP server or network manually entered Use the ping command at the os prompt (e.g. DOS prompt) to check step-by-step connectivity, from client station to access point and from access point to a wired network computer If 802. 1x authentication is in place, verify configuration and operation of the authentication server over a wired connection Some WLAN analyser products focus on one application area, such as spectrum or protocol analysis, while others combine these specific capabilities with more general performance and security analysis tools. The typical usage of these analysis tools is summarised in Table 9-6. With the advent of 802.11i and the increasing use of 802.1x authentication in WLANs, successful authentication becomes an additional step before a client station can successfully connect to the network. A WLAN analyser will be able to monitor each step of the EAP Chapter Nine Table 9-5: WLAN Troubleshooting - Checklist for Performance Problems Root cause Description Poor SNR - low signal Use a site survey tool (Section "Troubleshooting using strength WLAN analysers, p. 243") to test signal strength at the affected location Monitor signal strength while adjusting antenna location and orientation Consider increasing antenna gain or transmit power (up to regulatory limits) or relocating access points if signal strength remains low Poor SNR - high noise Use a site survey tool to identify other 802.11 transmis- level sions and non-802. 11 interfering signals Look for and eliminate the usual suspects (microwaves, cordless phones, Bluetooth) if high noise levels are identified Access point overload Survey users for any changes in applications or usage patterns Turn on and revi

ew the log for the access point experiencing performance problems A high re-try count under good SNR conditions will indicate re-tries due to competing traffic Consider additional access points on non-overlapping channels, or running dual mode networks (e.g. 802.11a and g) to increase capacity authentication process to see if a breakdown of this process is preventing user authentication and access. If the authentication server is denying a user access, the analyser results will help to determine whether the problem lies with the user's access rights or security configuration, or with the authentication server itself. Besides a wide range of commercial tools, a number of free or open source analysis tools are also available, the most popular being NetStumbler (see the Section "Wireless LAN Resources by Standard, p. 367"). This "free for non-commercial use" tool will identify SSID, channel, encryption settings and SNR of all detected access points, and can be used to verify network configuration and to detect any rogue stations within range. In small-scale implementations it can also be used to perform a simple site survey, checking for holes in RF coverage and detecting other 802.11a/g/b networks that may cause interference. Wireless LAN Troubleshooting Table 9-6: WLAN Analysers - Analysis Tools and Typical Usage WLAN analysis tools Typical usage Site surveying Locating non-802.11 interference sources Investigating intermittent connection problems caused by interference Monitoring of all 802.11a/b/g channels Determining a noise floor and identifying high noise or low SNR problems Verify access point channel usage and power level Mitigating channel overlap problems Performing pre-installation modelling and identifying problem areas with insufficient coverage Analysing site survey results in the light of specific technical requirements, e.g. for VoWLAN applications Security assessment Ensuring access points have policy compliant security con- figurations Visibility of encrypted network traffic (WEP, WPA, WPA2) Detec

FHSS systems, the 802.11 system will come off worse because its hopping rate is typically 160 times slower than the Bluetooth radio. Chapter Nine Channel 1 Channel 6 Channel 11 2.412 GHz 2.437 GHz 2.462 GHz 22 MHz 2.400 2.410 2.420 2.430 2.440 2.450 2.460 2.470 24 interfering FHSS channels per DSSS channel 79 X 1 MHz Bluetooth channels Figure 9-2: Bluetooth and 802.11 DSSS Spectrum Overlap This means that, in hopping over the 79 channels, the Bluetooth radio is likely to land on the same frequency as the 802.11 radio several times for each transmitted 802.11 packet. The 802.11 MAC will be issuing continued requests to repeat lost packets and network throughput will be degraded. Fortunately few 802.11 systems use the optional FHSS PHY layer specification. The situation is a little more complex for a DSSS 802.11 system (Figure 9-2), since the direct sequence detection is inherently more robust against narrow band interference, and as the probability of a collision between a FHSS packet and a DSSS packet depends on the WLAN data packet length. In this situation the Bluetooth link is likely to be more susceptible to interference since the DSSS interference will affect 24 of the 79 hopping channels, SO that some 30% of the WPAN packets could be lost. This will seriously degrade throughput, particularly for synchronous links such as voice transmission to a Bluetooth headset. The IEEE 802.15 Task Group TG2 has developed recommended practices to reduce the interference between 802.11 and 802.15.1 radios, using two types of coexistence mechanism - collaborative and non-collaborative. Collaborative mechanisms are possible when information to minimise interference can be exchanged between the WLAN and WPAN, while non-collaborative mechanism do not require exchange of information between the two networks, but are inherently less effective. The types of Wireless LAN Troubleshooting 802.11 beacon period Bluetooth traffic WLAN traffic Period timing WLAN interval WPAN interval TWLAN TWPAN Figure 9-3: WLAN and WPAN Transmit Perio

ds Defined in AWMA non-collaborative approach recommended are adaptive frequency hopping, adaptive packet selection and transmit power control. A collaborative TDMA mode termed alternating wireless medium access (AWMA) has also been recommended, where the available transmission time is divided between WLAN and WPAN transmissions, as shown in Figure 9-3. Because of the need for a communication link between the two networks, this collaborative mechanism can only operate if the two radios are located in a single host device - for example a laptop enabled for both Bluetooth and Wi-Fi. A further collaborative mechanism is termed deterministic frequency nulling. The concept here is to reduce the narrow band interference from the 1 MHz wide FHSS signal by nulling out this frequency at the 802.11b receiver. To do this the 802.11b receiver has to follow the hopping pattern and timing of the Bluetooth transmitter, and this is achieved by embedding a Bluetooth receiver within the 802.11b receiver. Although the Task Group is now officially in hibernation, a range of publications can be found at http://grouper.ieee.org/groups/802/15/pub. If Bluetooth interference is suspected as a cause of WLAN performance problems, an analyser such as AirMagnet's free BlueSweep utility can be used to identify active devices within the range of the WLAN operating area. Summary of Part III The implementation of a successful and secure wireless LAN requires attention to a broad range of issues. Chapter Nine Insight into the wide variety of technologies applied at the PHY and MAC layer in the different 802.1 WLAN standards provides the basis for understanding the technical capabilities of each standard and the reason behind their specific limitations, enabling the most appropriate technology to be selected for any WLAN application. An understanding of the basic RF propagation concepts, described in Part II, provides the background to the issues that need to be considered when selecting antenna equipment and planning the physical WLAN layout for

adequate RF coverage. Appreciation of the evolving technologies applied to secure WLANs enables appropriate choices to be made in deciding on the security requirements and on-going management implications for a particular WLAN implementation. A WLAN planned and implemented on the basis of a sound understanding of these underlying concepts and technologies will have the best possible chance of effectively meeting user performance expectations and security requirements. WIRELESS PAN IMPLEMENTATION Introduction A personal area network (PAN) is an interconnection of devices for personal use within the operating space of an individual - usually in the range of 1-10 metres. Wireless PANs aim to achieve this interconnectivity and give greater flexibility, mobility and freedom from the hassle of finding the right cable! WPANs differ from WLANs in that they are not intended to replace Ethernet type local networks, giving neither the range nor, at least at present, the data capacity or variety of services of WLANs. Instead they focus on the specific information and connectivity needs of the individual - synchronising data from a desktop computer to a portable device, exchanging data between portable devices and providing Internet connectivity for portable devices. Implementing successful WPANs is also less technically demanding than a typical wireless LAN implementation. Connections are generally quick to set up and call for little or no specific configuration. Chapter 10 looks at the technical and practical characteristics of a range of different PAN technologies, starting with the current de facto standard Bluetooth and including several emerging generic and niche competitors Part Four such as Wireless USB and ZigBee. The infrared IrDA standard, which has been very widely implemented as a flexible serial port replacement, is also covered. Implementation aspects including technology choices, security and other practical issues are covered in Chapter 11. CHAPTER Wireless PAN Standards Introduction Wires to connect the incr