Research On Single Station Passive Localization Based On Spatial Sparse Representation

Passive positioning can play a crucial role in battlefield reconnaissance,surveillance,and electronic warfare.This article focuses on the research of single-station passive location.For the location problem,because of the sparseness of the target location in the airspace,it is possible to apply compressive sensing in single-station passive location,which expands the method of single station passive location.Based on this,the main content of the article is as follows:First,the traditional single-station passive location method was studied.The weighted least square method based on angle of arrival,the method based on phase difference and its change rate,and the Taylor expansion method based on Doppler frequency difference are studied.In this paper,the implementation and simple derivation of these three methods are given,and the algorithm is simulated and analyzed.The influence of signal-to-noise ratio,number of positioning,measurement errors of correlation,etc.on the positioning is simulated.Then,the signal Doppler frequency difference and compressive sensing are combined in a single-station passive location.The Doppler frequency difference can be expressed as the linear sum of the pulses in the frequency domain.It is converted into the time domain to form a "pseudo-measurement value",and the result of sampling the "pseudo-measurement value" is taken as the observation value,then use sparse representation to construct an appropriate observation matrix,finally,the determination of the target position can be achieved by using a compressed sensing signal recovery algorithm.In this paper,simulation experiments are performed on this method.In addition,the relationship between the number of "pseudo-measurement" samples and the positioning error under different target numbers,the relationship between the Doppler frequency difference error and the positioning error are simulated.This method can easily locate targets when multiple target accurate Doppler frequency difference can be obtained.Finally,the signal covariance matrix and compressive sensing are combined in a single-station passive location.Different from the traditional method,this method does not need to first extract relevant observations such as DOA from the signal,it can directly use the covariance matrix of the sampled signal to obtain the positioning result in one step.In this paper,two different compressive sensing recovery algorithms are used to perform simulation experiments.In addition,different target numbers and positioning errors at different signal-to-noise ratios are simulated.However,with the increase in the number of targets,the performance of the positioning is slightly reduced,so an improved algorithm is subsequently presented.Firstly,the improved algorithm model and iterative steps are given.Then the performance of the improved algorithm under different target numbers and different signal-to-noise ratios is simulated and analyzed.In addition,under the same conditions,the simulation of the two algorithm models was compared.It was found that the improved algorithm still has good performance when the number of targets increases.Using the method of covariance matrix positioning can obtain the positions of multiple targets by simply processing the signals.The method is simple and the calculation accuracy is high.