plutoAp/Zhaohua_128_20260210DATA

128-Channel Acoustic Array — Partial UAV Signal Dataset (20260210)

Overview

This dataset contains partial UAV (Unmanned Aerial Vehicle) acoustic signals captured using a 128-channel acoustic imaging system on February 10, 2026. The dataset includes multi-session recordings with varying array configurations and source distances.

Note: This is a partial release. The complete dataset and source code will be made publicly available upon acceptance of the corresponding research paper.

Dataset Structure

The dataset is organized into 6 top-level directories, each corresponding to a recording session with naming convention: 20260210_<session>_<array>_<distance>

Directory	Array	Distance (m)	Audio (.wav)	Video (.mp4)
20260210_005_JinLing4_D30	JinLing4	30	192	8
20260210_009_JinLing4_D80	JinLing4	80	129	9
20260210_039_Matric4_D50	Matric4	50	193	8
20260210_041_Matric4_D80	Matric4	80	192	5
20260210_049_Matric4_D10	Matric4	10	192	2
BackGround	—	Background noise	128	0

Total: 1026 audio files (6.15 GB) + 31 video files (263 MB) = 1057 files, ~6.73 GB

Directory Layout (per session)

<session>/
└── 128ChData/
    ├── wav/
    │   ├── data-*.wav              # Full-array recordings
    │   └── split_output/
    │       └── 0.wav ~ 127.wav     # Per-channel split (128 channels)
    ├── *.mp4                        # Original and processed videos
    ├── final_calibrated_fusion/     # Calibrated fusion outputs
    ├── output_fusion_v1/            # Fusion method v1 outputs
    └── output_simple_fusion/        # Simple fusion outputs

File Descriptions

• wav/data-*.wav: Full recordings from the 128-channel microphone array (unsplit).
• wav/split_output/0.wav ~ 127.wav: Individual channel recordings after per-channel splitting.
• *.mp4: Video recordings including raw captures, YOLO-masked outputs, calibrated results, and motion detection results.
• final_calibrated_fusion/, output_fusion_v1/, output_simple_fusion/: Output videos from different signal fusion processing methods.

Background Data

BackGround/split_output/ contains 128-channel background noise recordings (no UAV target present), which can serve as baseline or negative samples for detection tasks.

Data Formats

• Audio: .wav (uncompressed PCM)
• Video: .mp4

Usage

This dataset is intended for research on:

• UAV acoustic detection and localization
• Microphone array signal processing
• Beamforming and sound source localization
• Audio-visual sensor fusion

Citation

If you use this dataset, please cite the corresponding paper (to be published).

Availability

• Current release: Partial dataset (audio + video only)
• Future release: Full dataset and source code will be uploaded upon paper acceptance