Research On Fast Optimization Design Methods Of Phase-coded Waveforms

Waveform optimization technology based on phase coding has become an effective means to improve the detection ability of weak radar targets.By designing different phase-coded waveforms for different application scenarios and platforms,it can effectively reduce the probability of radar being detected,suppress range sidelobes,and improve the performance of radar in complex environments such as strong clutter,interference,and multipath.With the continuous improvement of optimization theory,the traditional one-way waveform design has gradually begun to develop towards the cognitive waveform design with the environmental sensing ability and waveform online design capability.Therefore,studying fast phase-coded waveform design algorithms not only has important theoretical and practical significances for improving the detection performance of weak targets,but also provides theoretical and technical support for cognitive radar.In this dissertation,several design problems such as low sidelobe waveform design,orthogonal waveform set design,and sparse frequency waveform(set)design,are studied for single-input single-output(SISO)radar and multiple-input multiple-output(MIMO)radar.Chapter 1 first introduces the background and significance of the research on fast waveform design technology in radar,and then summarizes the research progress of phase-coded waveform design from two aspects of SISO radar and MIMO radar respectively.Finally,the content arrangement and organizational structure of this dissertation are given.Chapter 2 studies the design problem of low sidelobe waveform for SISO radar.Based on the peak sidelobe level(PSL)criterion,the design problems with unimodular constraint and peak-to-average ratio(PAR)constraint are established.For the problem with unimodular constraint,the gradient-based PSL minimization algorithm is proposed by deducing the phase gradient and the Newton downhill method used to obtain the iter-ative step size.For the problem with PAR constraint,the sequence synthesis method is used to transform it into an unconstrained minimization problem with respect to two unimodular sequences.Then,similar to the derivation process under unimodular constraint,the PAR-constrained gradient-based PSL minimization algorithm is proposed.Simula-tion experiments verify the effectiveness of the two algorithms,and also show that the waveforms designed under PAR constraint have lower PSL than the waveforms designed under unimodular constraint.Chapter 3 studies the problem of designing sparse frequency waveform(SFW)with the unimodular constraint for SISO radar.By weighting the stopband energy and the weighted integrated sidelobe level(WISL)criterion,the objective function of the SFW design problem is established.And then the waveform optimization algorithms based on the majorization-minimization(MM)framework and gradient are derived,respectively.The MM framework-based waveform optimization algorithm simplifies the objective function by applying the MM framework three times,and uses the acceleration strategy to improve the convergence speed of the algorithm;while the gradient-based waveform opti-mization algorithm uses the derivative rules and the Taylor series expansion to deduce the gradient and the step size,respectively.Moreover,in order to compare the convergence speeds of the algorithms,the MM process and the complexity are briefly analyzed.Simulation results show that the proposed algorithms can suppress the correlation sidelobes and the stopband power very well,and have faster convergence speed than the classical algorithm.Chapter 4 studies the SFW design problem with the PAR constraint for SISO radar.Compared with unimodular constraint,PAR constraint has more degrees of freedom,but it is also more difficult to handle.Firstly,for periodic and aperiodic waveform design,a unified optimization criterion is established in the frequency domain by the weighting method,and the PAR-constrained design problem is given.Then,this problem is transformed into an unconstrained minimization problem via the sequence synthesis method.Finally,by deriving the phase gradient matrix and the approximate step size,the PAR-constrained gradient-based iterative algorithm is proposed.Simulation results show that the proposed algorithm is superior to the existing algorithms in both waveform quality(the sidelobe and stopband performances)and computational efficiency.Chapter 5 studies the problem of designing orthogonal waveform set(OWS)for MIMO radar.Since the PSL criterion is more difficult to process than the integrated sidelobe level(ISL)criterion,there is no OWS design algorithm based on the PSL criterion.In order to simplify and process the PSL criterion of MIMO waveforms,the fp-norm criterion of MIMO radar is defined according to the autocorrelation sidelobe and the cross-correlation.Using the gradient method and the MM framework to optimize this lp-norm respectively,two efficient algorithms for OWS design are then proposed.Simulation re-sults show that the proposed two PSL-based optimization algorithms can achieve better sidelobe performance than the state-of-the-art ISL-based optimization algorithms.Chapter 6 studies the design problem of SFW set for MIMO radar.According to the WISL metric and stopband energy of the MIMO waveform,a unified optimization criterion that can be used for OWS design,complementary sets of sequences(CSS)design,and SFW set design is established.To minimize this criterion directly,the phase gradient matrix and the approximate step size are derived,and the gradient-based algorithm for designing SFW set is proposed.Then,for the design problem without considering the correlation weights,a simplified algorithm with low complexity is deduced.Simulation results show that compared with the existing MIMO waveform design algorithms,the proposed two algorithms can be applied to various waveform set designs,and have higher computational efficiency while ensuring waveform performance.Chapter 7 summarizes the research results and innovation points of the dissertation,and gives the future works.