Poster Abstract: Real-Time Active Identification and Tracking of UAVs Using Millimeter-Wave Radar

Unmanned aerial vehicle (UAV) poses a major threat to airspace security and privacy protection. However, existing UAV identification systems face the limitations of feature extraction quality degradation under low signal-to-noise ratio (SNR) conditions, and the average delay of traditional tracking algorithms is difficult to meet the real-time requirements due to high computational complexity. To address these issues, this paper proposes mmRTD, a real-time identification and tracking system for non-cooperative UAV sensing using mmWave radar. To solve the problem of serious feature extraction distortion in low SNR environment, this paper proposes a dynamic period adjustment mechanism driven by spectrum volatility to significantly improve the quality of feature extraction under low SNR conditions. To solve the problem of insufficient real-time performance of traditional tracking methods, this paper proposes a lightweight frame-based tracking framework through the target state detection method to solve the real-time bottleneck caused by high-dimensional signal processing of traditional methods. The experimental results show that the detection accuracy of mmRTD reaches 97.2% within a range of 40 m. In particular, the first detection time of mmRTD is two orders of magnitude higher than that of the traditional scheme, which verifies its potential for real-time identification in complex scenes.