Research On Optimization Algorithms For MIMO Radar Waveform Design

MIMO radar can transmit independent signals,which brings the advantage of waveform diversity.It has significant advantages in terms of system freedom,signal processing at the receiving end,and interference suppression performance.Therefore,MIMO radar can adapt to complex battlefield situations and communication scenarios,and plays an increasingly important role in both the national defense and civilian fields.In recent years,related research on MIMO radar has gradually become a research hotspot in the field of radar.Among them,the optimized design of the MIMO radar transmit waveform can effectively improve the target parameter monitoring and tracking performance of the MIMO radar,which is a difficult problem that needs to be solved urgently in related research.This thesis investigates the related algorithms of MIMO radar waveform design.The main work is as follows:1.Establish the signal model of MIMO radar.Then summarize several common criteria for MIMO radar waveform design and introduce the main optimization methods of waveform design.2.Research on the waveform optimization design problem based on MIMO radar transmit beampattern matching.First,the original bivariate target problem is transformed to a univariate problem.Then the residual network is constructed and the waveform is optimized by the deep learning method,which further improves the matching degree with the expected emission pattern.3.The MIMO radar waveform optimization problem based on maximizing SINR is investigated and two optimization methods are proposed: the optimization method based on convex relaxation,the optimization method based on Riemannian manifold.In the optimization method based on convex relaxation,the original target problem is reconstructed first.Then an auxiliary variable is introduced to transform the non-convex optimization problem into a convex optimization problem.Finally,the waveform is optimized by alternating optimization.In the optimization method based on Riemannian manifolds,the original problem is modeled as an unconstrained optimization problem on Riemannian complex circle manifold,and the Riemannian conjugate gradient method is applied to solve it iteratively.Both methods can achieve higher SINR.4.In this section,the optimization problem for MIMO radar including transmit waveform optimization and the receive end optimization.In the optimization,the optimization criterion is to maximize SINR,and the optimization of the transmit and the receive end are performed separately.First,the Riemannian manifold optimization method is applied to optimize the transmitted waveform.Then MVDR algorithm is used to optimize the filter weight vector.The transmit beampattern obtained by the proposed method can generate a deeper notch in the interference direction while obtaining a high SINR.