Radar anti-deception system for identifying ghost objects resulting from reciprocity-based sensor deception
By using time-matched cluster computation and threshold technology from radar sensors to identify ghost objects, the problem of reciprocity deception in radar sensors is solved, and a highly efficient ghost object identification effect is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GM GLOBAL TECHNOLOGY OPERATIONS LLC
- Filing Date
- 2022-10-25
- Publication Date
- 2026-07-21
AI Technical Summary
Existing anti-spoofing technologies struggle to effectively identify ghost objects generated by reciprocity-based sensor spoofing, especially in radar sensors.
Input detection points generated by multiple radar sensors are used to determine time-matched clusters by executing commands from the controller. The adjusted signal-to-noise ratio (SNR) and velocity ratio are calculated, and ghost objects are identified by thresholding techniques. Object classification is performed by combining fuzzy c-means clustering algorithm and nearest neighbor technique.
It achieves efficient identification of ghost objects generated by reciprocity-based sensor deception, with an average classification accuracy of 92.3%, effectively mitigating the impact of radar sensor deception.
Smart Images

Figure CN116263492B_ABST