Design Of MIMO Radar Waveform Based On Ambiguity Function

With the gradual development and perfection of Multiple-Input Multiple-Output(MIMO)technology,the demand for cross-application of MIMO technology also shows an exponential growth.As a result,many researchers consider the wide application of MIMO technology,MIMO radar technology emerged as the times require.In radar system,signal design and optimization is a core problem of system design,and it is very important to the performance of radar system.The ambiguity function of the radar waveform is an effective mathematical tool to estimate the target range and velocity.Therefore,this paper designs radar signal waveform based on ambiguity function.In the process of waveform design,the distance between covariance matrix and expectation matrix is usually the smallest.Euclidean distance is often used in existing distance measurement methods,but considering the structural characteristics of covariance matrix,the accuracy of Euclidean distance measurement is low.At the same time,in the process of signal optimization,because the variable design problem is a quadratic non-convex function,the optimal solution can not be obtained directly.Based on the above problems,the main work of this paper is as follows:1.In order to minimize the distance between covariance matrix and expectation matrix in signal design,a signal design method based on Riemann distance is proposed in this paper.The objective of this paper is to design a signal vector with expected covariance matrix and to ensure that the side lobe of the ambiguity function of the signal is small.Because covariance matrices are structurally constrained,they form manifolds in the signal space.Therefore,the distance between these matrices should not be measured by Euclidean distance,but by Riemann distance along the surface of manifold.2.Aiming at the problem of quadratic non-convex function in signal optimization,an efficient algorithm based on successive convex quadratic optimization is developed.The algorithm transforms the quadratic optimization problem into quadratic convex optimization problem,and then uses CVX toolbox to solve it.By comparing the signal design based on Euclidean distance and Riemann distance,the experimental results show that the convergence speed of the objective function based on Riemann distance is obviously faster than that based on Euclidean distance.3.In order to verify the effect of the designed signal on the performance of radar system,we use the optimized signal to estimate the range and velocity parameters of co-located targets.Compared with the error estimation results under the Euclidean distance and Riemannian distance optimization methods,we find that when the target spacing is relatively close,the error yielded by signals designed under the Riemannian distance optimization is smaller,thus,increasing the accuracy of the estimation performance of the radar system.