Research On Resource Allocation Approach For Target Localization And Tracking With Distributed MIMO Radar

MIMO radar is typical of new radar system,and it has become an exciting technology.In comparison to collocated MIMO radar,distributed MIMO radar has a greater breakthrough on traditional radar system.It can observe the target from different angle,which provide it with advantage on overcoming the radar RCS fluctuation and anti-stealthtechnology.The research for resource allocation in distributed MIMO radar system has become a hotspot to explore its ability of spatial parallel multi-channel access to information further.In this paper,the joint resource allocation approach for improving target localization and tracking performance with distributed MIMO radar system is studied.The main work includes:1.The factors that have influence on target squared position error bound(SPEB)are analyzed,and a joint resource allocation scheme incorporating system sensor,power and signal bandwidth is proposed for better target localization performance.Firstly,a criterion minimizing the SPEB on the mean square error in target location estimation is derived,and the corresponding optimization model with transmitted sensors,power and signal bandwidth is simplified by deriving the relationship between the optimal power allocation vector and bandwidth allocation vector.Then,the optimization problem is transformed into a second-order cone programs(SOCP)by convex relaxation and cyclic minimization algorithm(CMA).Finally,the problem can be solved by interior-point methods.The simulation results demonstrate that the localization accuracy is improved by the proposed algorithm.2.The target velocity estimation accuracy is related with the system sensor,transmitted power and signal duration time when the transimiting signal satisfy achieve orthogonality,and a joint resource allocation algorithm is proposed for improving the velocity estimation accuracy for multiple targets.Firstly,a criterion minimizing the maximum Cramer Rao Lower Bound(CRLB)on the mean square error in multiple targets velocity estimation is derived,and the corresponding optimization model is built.The non-convex optimization problem is transformed into a sequential convex optimization problem by Sequential Parametric Convex Approximation(SPCA)algorithm.Finally,simulations demonstrate that the velocity estimation accuracy is improved by the proposed algorithm.The results also reveal that signal effective time width has a greater impact on the velocity estimation accuracy than transmitted power.3.Firstly,the model for target tracking in MIMO radar is set up.Then,the Bayesian Cramer Rao lower bound(BCRLB)on the mean square error in target location and velocity estimation is derived by system observation and Intel Mobile Module-Unscented Kalman Filte(IMM-UKF).And the tracking accuracy is related with the transmitted power,signal bandwidth and signal effective time width by analyzing the BCRLB.Therefore,a joint resource allocation algorithm to improve the maneuvering target tracking accuracy is proposed.Firstly,a criterion minimizing the BCRLB is derived,and the corresponding optimization model with these four resource variables is constructed.To solve this non-convex problem,we adopt cyclic minimization algorithm and convex relaxation approach to transform it into a convex optimization problem.Finally,simulations demonstrate that the target tracking accuracy is improved by the proposed algorithm.