Abstract

Radar signal sorting is a key component in electronic support measures (ESM). Adaptive sorting algorithms are a type of method that does not require prior understanding of radar information in the sorting scenario. However, these methods have consistently faced the issue of high runtime costs. To address this problem, in this paper, a dual-mode radar signal sorting method based on density clustering is presented. Adaptive density peak clustering (ADPC) is used as the initial algorithm for sorting and extracting radar information from the sorting results, which is defined as prior knowledge. The prior knowledge is then passed to Gaussian mixture model (GMM) to facilitate a more rapid sorting process. When changes in the radar within the sorting scenario are detected, it switches back to ADPC. This method allows for rapid sorting of stable data while promptly generating new prior knowledge in response to data changes. Numerical experimental results demonstrate that the proposed method achieves high precision in sorting results while using a lower time cost compared with the prevailing sorting methods.