Wavelet Packet and Granular Computing With Application to Communication Emitter Recognition

Individual recognition of communication emitter can obtain the working mechanism, system parameters, and the usage model of the emitters based on the separation and interception of the specific emitter signals. In order to realize the individual recognition of communication emitter, a method of communication emitter recognition based on wavelet packet feature extraction and attribute reduction of granular computing is proposed. The method uses hierarchical and granularity to represent the subtle feature information of different emitter signals contained in the original data after the decomposition of wavelet packet, so that the feature attributes of different emitters are efficiently mapped to different levels of granular structure which can be used to search for an optimal reduced attributes operation for signal recognition. And then, the reduced attributes are used to distinguish the emitter signals correctly by label propagation method. The experimental results show that the proposed method based on wavelet packet feature extraction and granular computing attribute reduction has good classification performance in individual recognition of communication emitters.