Broadband Radar Echo Target Parameter Estimation Based On Orthogonal Compressed Sampling

With the increasing demand for "refined" radar information,researchers pay more and more attention to the research of wideband radar.Wideband radar has advantages over narrowband radar in terms of radar resolution,target feature recognition and anti-jamming due to its large bandwidth.However,there are two important issues they have to be faced with.The first one is that due to the increasing of signal bandwidth,the traditional ADC technology based on the Nyquist sampling theorem will be confronted with technical bottleneck,and the increasing of the amount of sampled data also brings difficulties to the subsequent processing.Another problem is that due to the increase of the transmitted signal bandwidth,the approximate condition of the narrow-band model no longer holds.The scale conversion effect causes the center frequency and bandwidth of the echoes to change with respect to the transmitted signal.The traditional range-Doppler processing is no longer applicable in wideband radar.In recent years,the theory of compressed sensing provides a new method for effectively acquiring wideband radar signals.Compressive sensing theory is a new signal sampling theory based on signal sparse representation and convex optimization theory.It takes the signal information rate(sparsity level)as the criterion and could acquires the signal far below the Nyquist sampling rate.Based on the compressive sensing theory,this paper studies the compressive sampling and parameters estimation of wideband radar echo signals.In this paper,the wideband radar echo signals are acquired at low rate by utilizing the quadrature compressive sampling system.Then the wideband waveform matching dictionary is used to sparsely represent the wideband radar echo signals.Finally,the reconstruction of the wideband radar echo signals and the joint estimation of time delay and scale factor is realized relying on the sparsity of signals.The main work of this paper is organized as follows:(1)Briefly describe the basic theory of compressed sensing and analog to information conversion.First of all,we briefly introduce the sparse representation of signals,the compressive measurements of sparse signal and signal reconstruction.Then,we introduce several typical analog to information conversion systems.Finally,the quadrature compressive sampling analog-information conversion system for radar applications proposed by my research group was introduced in detail.(2)Develop the method of wideband radar echo signal reconstruction and parameter estimation based on quadrature compressive sampling.In this paper,the quadrature compressive sampling system is applied to wideband radar signal acquisition to effectively achieve low-speed sampling of wideband radar echo signals.In order to reconstruct the original signal from the low-speed sampled data,we propose a wideband waveform matching dictionary based on time delay and scale factor two-dimensional joint discretization and then sparsely represent the wideband radar echo signal.Utilizing signal sparsity,wideband echo signals can be reconstructed by sparse reconstruction algorithms and target parameters can be estimated on this basis.The paper simulates and analyzes the reconstruction performance and the target parameter estimation performance of the method,and the simulation results verify the feasibility and effectiveness of the method.(3)A sparse Bayesian learning based time delay and scale factor joint estimation method is developed.Because the correlation of wideband waveform matching dictionary based on two-dimensional discretization of time delay and scale factor is relatively high when the discretization grid is fine,the performance of sparse reconstruction may be affected.In order to effectively improve the performance of joint estimation of time delay and scale factor of wideband echo signals,this paper chooses to apply the sparse Bayesian learning to parameters estimation of wideband radar echo signals.A method of time delay and scale factor joint estimation based on sparse Bayesian learning is proposed to effectively improve the performance of parameter estimation of wideband echo signals.The simulation results show that the proposed method can effectively improve the accuracy of the target parameter estimation and anti-noise capability.