Investigation On Transmitted Waveform Design And Optimization For Modern Radar System

Recently,several techniques such as cognitive radar,multiple input multiple out-put(MIMO)radar,compressed sensing radar(CSR),etc.greatly promoted the design and development of radar system and bring a revolution of radar technology.The con-cept of cognitive radar connects the original isolated transmitter and receiver design and realizes the closed loop of the joint transmit waveform and receive filter design of the radar system.MIMO radar introduced the design method of high dimension-al system for space diversity and waveform diversity,which expanded a new degree of freedom for radar system design.The cost of these advantages is the pressure of acquisition,transmission and processing of multi-dimensional data.The theory of compressed sensing provides a solution to this problem.It breaks through the limi-tation of the traditional Nyquist sampling theorem,and accomplishes the compressed sampling and reconstruction of signals based on sparsity or compressibility of interest-ed signals.CS theory provides a new approach to solve the problems of the realization and processing of radar system under the mass data.As an important part of the radar system design,this thesis centers on radar waveform design and beamforming for these modern radar system.The main works are illustrated as follows:1.Wideband sparse spectrum waveform designTo deal with the problem of the narrow-band interference mitigation faced by wide-band and ultra-wide-band(UWB)radar,studies on the adaptive optimization design of sparse spectrum waveform is carried out.For reference by using the method of transmitted waveform design to suppress the interference of narrow-band interfer-ence in the wireless communication system,a new objective function is proposed to optimal transmitted waveform of radar system.The proposed function can flexibly constrain the pass-band and stop-band performance of the transmitted waveform ac-cording to the external environment,and match the power spectrum density(PSD)of the transmitted waveform with the expected PSD,in order to optimize the anti-interference characteristic of radar system.At the same time,the integral sidelobe level(ISL)is constrained.The optimal weight of PSD and ISL can be balanced ac-cording to the actual application requirements.Then,based on the assumption of unimodular sequences and combined with the Taylor expansion of the objective func-tion,a cyclic phase-only conjugate gradient method is proposed to get the optimal phase modulation waveform at a very fast convergence speed.Optimized waveform can effectively suppress narrow-band interference,have low ISL and aviod the loss of signal to interference and noise ratio(SINR)and main lobe broadening caused by mis-match processing.All these advantages determine that the optimized waveform has a strong ability to adapt to the bat,tlefield environment.2.Single-stage waveform synthesis for colocated MIMO radar beampattern match-ingMIMO radar beamforming algorithms usually consist of a signal covariance ma-trix synthesis stage,followed by signal synthesis to fit the obtained covariance matrix.In order to realize the MIMO radar beamforming more simply and efficiently,a MIMO radar beamforming design framework is proposed.This framework does not need to consider the emission autocorrelation matrix in the design process;i.e.no prior co-variance matrix synthesis is required.According to proposed framework,the approx-imation and system loss introduced by the step optimization are overcome.Further more,the scale of the optimization problem can be reduced by reducing the dimension of the objective function.Based on the proposed design framework,a MIMO radar transmitted waveform optimization algorithm using SVD decomposition is presented.This optimization algorithm can optimize the transmitted waveform of MIMO radar to fit the ideal beampattern.Then,in order to avoid the huge computation and the matrix decomposition operation which is difficult to achieve in practical engineering,a Beam-Shape algorithm based on cyclic iteration for single step MIMO radar beampat-tern matching is proposed.The monotonicity of the iterative Beam-Shape algorithm is proved,and the convergence of the algorithm is guaranteed.Subsequently,the pro-posed Beam-Shape algorithm is further extended from the design weighting matrix to the direct optimization design of the MIMO radar transmitting waveform,so as to cope with the constraint of the transmitting waveform of single array in practical ap-plication.Finally,the common sets of transmitted waveform constraint are discussed,and the projection method under different constraint sets is given.3.CSR waveform design based on range-Doppler parameters estimationAccording to the framework of CSR model,transmitted waveform design for both basic single-pulse and general pulse-train scenarios with range-Doppler parameters es-timation is studied.First,the range-Doppler parameters estimation model of general pulse-train CSR system is established.Note when the number of transmitted pulses is equal to one,the model can be degraded to single-pulse CSR model.Through the analysis of the model,it is pointed out that the echo of CSR in pulse train system can be expressed as a function of target scattering vector.Since the interested targets only occupy a small amount of radar resolution cells,the received signal of the targets within the radiation region is sparse and CS theory can be used to estimate the per-ception model.Followed this idea,the objective function of CSR sensing matrix on transmitted waveform is derived.Then two cyclic transmitted waveform optimization algorithms for both basic single-pulse and general pulse-train scenarios are presented respectively.Finally,we analyze and evaluate the proposed algorithm from aspects of distance and speed detection performance,target scene recovery probability,parameter estimation error and computation complexity.Simulation results show the effectiveness of the proposed method.