A novel deinterleaving method for radar pulse trains using pulse descriptor word dot matrix images and cascade‐recurrent loop network

Traditional signal deinterleaving methods depend on preset parameters, use complicated steps and have limitations in handling complex electromagnetic environments where signals overlap the space, time frequency domains. This paper presents a novel approach using deep segmentation network for radar deinterleaving. The pulse descriptor word data is transformed into dot matrix image, which then processed by Cascade-Recurrent Loop Network (CRLN) segmentation. CRLN consists of Type Segmentation Network, an Amount Decision Recurrent Individual Network. By recursively segmenting each target determining number sources, overcomes challenges posed discontinuity high individuals image. Experimental results demonstrate effectiveness proposed method, surpassing traditional methods, achieving accuracy, low omission rate effectively mitigates increasing batch problem scenarios. experiments conducted randomly generated dataset yielded impressive results: over 90% accuracy when dealing with pulses from 15 radars different types, 96% 4 same type 5 types 10% loss.