角闪烁噪声下的集中式MIMO雷达自适应资源分配算法

Adaptive resource allocation algorithm in collocated MIMO radar under angular glint noise

为解决角闪烁噪声下集中式多输入多输出(multi-input and multi-output,MIMO)雷达的资源优化分配问题,设计了一种面向多目标跟踪任务的自适应资源分配算法。首先采用平方根容积粒子滤波(square-root cubature particle filter,SCPF)算法对各目标状态进行估计,并根据其估计值来计算条件后验克拉美罗下界,从而建立起角闪烁噪声下的跟踪误差评价准则。再依据功率和带宽与条件后验克拉美罗下界(predicted conditional Cramer Rao lower bound,PC-CRLB)间的函数关系,建立起非凸优化模型。随后运用凸松弛技术和循环最小化算法将非凸优化问题转换为一系列凸优化问题,运用半正定规划(semi-definite programming,SDP)算法结合Frank-Wolfe可行方向法进行求解,从而实现资源自适应分配。最后通过仿真验证了不同场景下所提算法的有效性。

To solve the resource allocation problem in collocated multi-input and multi-output(MIMO)radar under angular glint noise,an adaptive resource allocation algorithm for multi-target tracking tasks is designed.Firstly,the square-root cubature particle filter(SCPF)algorithm is adopted to estimate the state of each target,and then the predicted conditional Cramer Rao lower bound(PC-CRLB)is calculated based on its estimate value,thus the evaluation criterion of tracking error under angular glint noise is established.Then a non-convex optimization model is established based on the function relationship of power and bandwidth with PC-CRLB.In addition,the non-convex optimization is transformed into a variety of convex optimization problems by using convex relaxation technique and cyclic minimization algorithm,and the semi-definite programming(SDP)algorithm combined with the Frank-Wolfe feasible direction method is used to solve the problems,so as to realize the adaptive allocation of resources.Finally,the effectiveness of the proposed algorithm is verified by simulations under different scenarios.