Abstract: As a typical "low, small, slow" target, UAV has the characteristics of slow flight speed, low altitude, and small radar reflection area (RCS), making it difficult to detect and identify UAV targets. In view of the problem of low signal-to-noise ratio and difficult detection of UAVs in complex environments, a knowledge-aided constant false alarm rate (CFAR) detection method for UAV targets is proposed. This method first analyzes three common ground clutter distribution models and mean CFAR detectors, and then adopts CFAR detection methods for the echo signals under the three clutter distributions, and uses the method with the best detection performance as the clutter distribution The optimal CFAR detection method is stored in the knowledge base to establish the CFAR knowledge base; by estimating the clutter distribution of the echo signal of the target to be detected, the clutter distribution model is judged, and the distribution is obtained from the radar knowledge base Select the corresponding CFAR algorithm to complete the echo signal detection. Finally, the actual measurement data collected by radar is used to verify the feasibility and effectiveness of the method.