Fast Algorithms For Extracting Main Ridge Slice Of Ambiguity Function Of Radar Emitter Signals

Radar emitter signal sorting is a key procedure of signal processing in modern high-technology war and future information war.The classical methods of pulse trains deinterleaving are generally based on five classical parameters.Now,as the electronic countermeasure is becoming more and more intense, the deinterleaving of radar emitter signal based on five parameters already becomes more and more impractical,so to explore and supplement the new effective sort characteristic parameter is a feasible method to solves this question.Considering the ambiguity function can provide the specific effect of a signal inherent structure,PU Yunwei proposes a novel approach for searching the parameters of main ridge slice of ambiguity function of radar emitter signals to sort the signals.Experimental results show that the features of main ridge slice have strong compactness in a cluster and good separateness between cluster, its performance of anti-noise is good. If want get a precision result of main ridge slice of ambiguity function,it requires huge computational complexity,so it is disadvantage to practical application.Therefore, the processing problem that how to quickly and accurately extract the main ridge slice of ambiguity function in fraction domain is researched, the main work and research are as follows:1. Recent researches and developments of signal sorting of radar emitter are brief introduced.The principle of an approach which is based on fractional autocorrelation to extract the main ridge slice of ambiguity function is introduced.Using discrete-method in fraction domain to resarch main ridge slice of radar emitter signals and give a specific algorithm flow. Experiments conducting on radar emitter signals show that it require huge computational complexity to get a precision result.2. A theoretical analysis about the ambiguity function of LFM signal,a conclusion is obtained that the ambiguity function of LFM signal is symmetrical,unilateral and strictly monotone.Based on this conclusions new method(iteration-menthod) is proposed to search the main ridge slice of LFM signals. Compare with discrete-method,Experiments results show that this approach have the advantages of fast searching speed and precise result.3. The iteration refer only to LFM signal, so,its generality is not good. Considering that the quantum evolutionary algorithms(QEA)is characteried by small population slice,rapid convergence and strong global searching capability, so,it transform the problem of searching the main ridge slice of ambiguity function into a optimization problem and propose a new approach that based on QEA to extract the main ridge slice of ambiguity function and give a specific algorithm flow. Experimental results show that the computational amount reduced enormously without lowering the precision of the parameter for a variety of signals.This paper was supported by the National Natural Science Foundation of China (Grand No.60702026) and Youth Science and Technology Foundation of Sichuan (09ZQ026-040).