Research On Signal Reconstruction Algorithm Of Compressed Sensing Radar Receiver Signal Basing On AIC

Wide-band radar signal plays an important role in radar technology because of its high range resolution and good anti-jamming performance.When acquiring wide-band radar signals without distortion basing on the traditional Nyquist sampling theorem,the sampling rate needs to be twice as high as the highest frequency of the signals.Due to sampling jitter and bandwidth limitation,the existing digital-to-analog converter(ADC)can no longer meet the demand of high-speed sampling with the continuous expansion of broadband.On the other hand,a large number of sampling data obtained by conventional sampling also put forward higher demands on the transmission,processing and storage of digital signal processing chip.The new theory of compressed sensing(CS)breaks the limitation of Nyquist theory on information acquisition.Based on the sparsity or compressibility of the signal,the CS uses an observation matrix to map the signal directly from the high-dimensional space to the low-dimensional space.Observation matrix is not related to the sparse basis or dictionary.The sparse constrained optimization algorithm is used to reconstruct the signal basing on the few observations.In order to solve the problems of poor sparsity in frequency domain and low reconstruction precision of wide-band radar signals,the redundant dictionary formed by the cascade of signal level dictionaries based on radar signals' own characteristics and DFT orthogonal bases is regarded as sparse dictionary of signals in this paper.Based on the knowledge of block sparse signal,this paper proposed a fast signal reconstruction algorithm-cascading dictionary matching pursuit algorithm based on block-sparsity(BS-CDMP).BS-CDMP algorithm can recognize target information directly.Combined with the main research content of this paper,BS-CDMP algorithm,a radar receiver framework based on piecewise integrated analog-information-converter(AIC)structure was presented.This framework can directly and efficiently compress and sample the analog radar signal,and reconstruct the signal according to the compressed sampling value.Among them,the emphasis of the research is to use BS-CDMP algorithm to complete the signal reconstruction under this framework.In the process of reconstruction,the objects in the range of interest can be detected and reconstructed quickly and accurately;At the same time,it also has high detection and reconstruction ability for non-interested signals.The concrete results of this paper are as follows:(1)Three feasible AIC structures are analyzed,that is,pseudo-random demodulation single branch AIC,parallel AIC and piece-wise integral AIC.The physical and mathematical models of each structure are given.The discretization matrix representation corresponding to each model is further analyzed.Three kinds of AIC structures are simulated by using matrix representation on Matlab platform and performance comparison with each other.(2)The sparse representation of radar signal is studied.Aiming at the problem of poor sparse representation of LFM radar signal based on common DFT orthogonal basis.The signal level dictionary is constructed according to the characteristics of signal,and the sparse expression ability of the signal level dictionary is verified.Aiming at the problem that the single orthogonal basis is not universal for radar signal,the cascade redundant dictionary,which cascading the signal level dictionary with direct signal information with the DFT orthogonal basis,is regard as the sparse dictionary of the signal.The condition of optimal sparse representation of signals in cascade redundant dictionary is proved.Based on block sparsity theory and cascade dictionary sparse representation method,this paper proposed a fast signal reconstruction algorithm-cascading dictionary matching pursuit algorithm based on block-sparsity(BS-CDMP),which can quickly recognize and reconstruct signals.The concrete steps of the proposed algorithm are given and the LFM signal is used for simulation.(3)Combining the piece-wise integral AIC structure with the proposed BS-CDMP reconstruction algorithm,a framework for compression sampling and reconstruction of analog signals of radar receivers is proposed.The performance of under-sampling and reconstruction of radar signal of the proposed framework is verified on Matlab platform.And the performance is analyzed by simulation results.