Spare Ultra Wideband Signal Analysis In Electronic Reconnaissance

In the modern electronic warfare,as the range of radar reconnaissance frequency is increasing,the analysis of sparse UWB electronic reconnaissance signals is particularly important.The traditional Nyquist sampling theorem has become a bottleneck restricting the development of broadband signals.Compressive sensing theory brings a dawn to the development of this technology.It points out that when the signal is sparse in a certain domain,it can break through the Nyquist sampling rate,and then use the recovery algorithm to accurately reconstruct the signal.Therefore,how to use the compressed sensing theory to accurately estimate the parameters of the target signal has become a hot topic in the field of signal processing.In this paper,Combinatorial Compressive Sensing is used to study the above-mentioned problems.The main research contents are as follows:1.Apply the Bayesian framework's fast Bayesian matching pursuit reconstruction algorithm to signal reconstruction in electronic reconnaissance,and compare it with compressed sampling matching tracking reconstruction algorithm under the framework of compressed sensing theory.The reconstruction accuracy of the fast Bayesian matching pursuit algorithm is higher than that of the compressed sample matching pursuit algorithm.Under the same conditions,the FBMP algorithm can reconstruct the signal at a lower signal-to-noise ratio.2.The parameter estimation based on the waveform matching dictionary is studied.The redundant dictionary is constructed according to the waveform characteristics of multi-frequency sinusoidal signal and chirp signal,and then the frequency parameters of the signal are estimated by the solution optimization algorithm in compressive sensing.The experimental simulation shows that the algorithm can accurately estimate the frequency of the signal.Information,but when constructing a dictionary,we must know the prior information of the frequency range of the signal,and the constructed dictionary is also very large,making the computational complexity very high;3.For wide-band signals,we combine the fractional Fourier transform with the compressed sensing framework and construct a fractional Fourier dictionary to estimate the frequency of the LFM signal.Firstly,the two-dimensional search is used to build the fractional Fourier dictionary of each order p according to the length of the non-synchronization,and then the optimal order p and the parameter u are selected by the optimization algorithm in the pressure sense,and the starting frequency of the signal is determined by the formula.The slope of the FM is estimated.The simulation results show that the estimation error of the algorithm to the FM slope can reach the order of magnitude,and the estimation error of the initial frequency can reach the order of magnitude;4.Use the signal source to generate various types of signals,perform compression sampling through the existing non-uniform sampling platform,obtain the measured data,and perform multi-frequency sinusoidal signals and single-frequency pulses through compressed sample matching tracking and fast Bayesian matching pursuit algorithm The signal and chirp signals were spectrally reconstructed,which further validated the feasibility of the compressed sample matching pursuit and the fast Bayesian matching pursuit algorithm.The reconstruction effects of these two algorithms were analyzed and compared.Then the fractional Fourier dictionary is used to estimate the frequency of the measured data,and a good result is obtained,which verifies the effectiveness of the method.