A Research Of Radar Target Localization Based On Compressed Sensing

Radar is an important fighting weapon in modern military. Target localization isits basic and also the most important function. According to the characteristics ofMIMO radar, the Direction-of-arrival (DOA) parameter estimation of traditional arrayradar is applied to the MIMO radar target localization. Related Analysis shows thatthe traditional radar target localization methods make the large signal scale of MIMOradar becoming larger, greatly increasing the difficulty of data processing. In fact, thetargets relative to the background are highly sparse, which makes the extremeimbalance in huge amounts of data collected from radar system and the sparsity oftargets. Compressed sensing theory effectively used the sparsity or compressibility ofsignals, it breaks the bottleneck of the traditional Nyquist sampling theorem andmakes the sampled data much less than the amount of conventional method’ssamplings, greatly reducing the signal sampling and transmission costs.In this paper, a new radar target localization method based on compressedsensing is proposed for lack of the high sampling rate in traditional radar targetlocalization method. The main works are as follows:Firstly, the thesis analyzes the status quo of the traditional radar targetlocalization and the application status and prospects of compressed sensing theory inthe field of radar. It describes the framework of compressed sensing theory in detail,and conductes a focused analysis in comparing the performance of several commonreconstruction algorithms.Secondly, the thesis presents the principles of MIMO radar and signalcharacteristics, establishes a signal model of MIMO radar converting the problem ofMIMO radar target localization to sparse matrix reconstruction in compressed sensingtheory. According to the MIMO radar target localization which roots in the basispursuit(BP)algorithm, the thesis proposes a new target localization method based onorthogonal matching pursuit(OMP)algorithm. Simulation results confirm theeffectiveness of this method.Finally, for the problem of the MIMO radar target localization based on OMPalgorithm need to know the targets’ number in advance, the thesis proposes a MIMOradar target localization based on sparse adaptive matching pursuit (SAMP) algorithm.Simulation experiments are conducted to compare the localization performance of the three methods. Simulation results show that: the method of MIMO radar targetlocalization based on SAMP algorithm has a better performance.