Recognition of radar emitter signals using spectrum atoms and Hierarchical Decision Strategy

An effective approach to recognize the radar emitter signals is presented, which is based on a novel Spectrum atoms and Hierarchical Decision Strategy. Firstly, a new over-complete multi-scale Spectrum atom dictionary is constructed and the improved MP algorithm is applied to extract the features of radar emitter signals, the optimization of atom dictionary structure and FFT is introduced to effectively reduce the time complexity at each search step of MP. Then, some intrinsic Spectrum atoms describing features of signals are obtained which has the good properties of clustering the same and separating the different kind of radar emitter signals. Finally, a hierarchical decision strategy is exploited to identify signals automatically. Simulation results denote that the proposed approach have good performances of recognition with low SNR, which confirms the validity and feasibility of the approach.