非准确先验知识下认知雷达低峰均比稳健波形设计  

Low-PAR Robust Waveform Design for Cognitive Radar with Imprecise Prior Knowledge

作　　者：郝天铎 周青松[1] 孙从易 崔琛[1] 

机构地区：[1]电子工程学院,合肥230037

出　　处：《电子与信息学报》2018年第3期532-540,共9页Journal of Electronics & Information Technology

摘　　要：针对目标和杂波先验知识不准确时认知雷达的检测波形设计问题,同时兼顾功率放大器对低峰均比(PAR)波形的需求,该文提出一种信号相关杂波背景下认知雷达低PAR稳健波形设计方法。首先,在目标和杂波不确定集范围内,基于极大极小化准则构造关于输出信干噪比(SINR)的优化模型;然后将不确定性参数代入该模型,给出最差SINR下对应杂波协方差矩阵和目标Toeplitz矩阵的取值;在此基础上,利用半正定松弛,将非凸的优化模型转化为关于发射波形半正定矩阵的凸问题进行求解;最后,通过秩1近似法结合最近邻方法,进一步从波形的最优矩阵解中提取出最优向量解。分析表明,在稳健性能相同的情况下,与现有方法相比该算法具有更低的运算量,仿真结果验证了所提方法的有效性和稳健性。In view of the detection waveform design for cognitive radar with imprecise prior knowledge of target and clutter, while considering the demand of power amplifier on low Peak-to-Average power Ratio（PAR） waveform, a low-PAR robust waveform design method in presence of signal-dependent clutter is proposed. Firstly, the optimization model of radar＇s output Signal-to-Interference-plus-Noise Ratio（SINR） is established within the uncertainty of target and clutter via Max-Min method. Secondly, the clutter covariance matrix and Toeplitz matrix of target corresponding to worst-case SINR is obtained. Since the optimization problem of waveform is non-convex, Semi-Definite Relaxation（SDR） is adopted to converse the non-convex problem into a convex problem, which is about the semi-definite matrix of waveform. Finally, the optimal vector solution of waveform can be extracted from the optimal matrix solution by the rank-one approximation method combined with the nearest neighbor method. Compared with the existing methods, the computational complexity of the proposed method is obviously reduced without losing robust performance according to the analysis. The simulation results demonstrate the effectiveness and robustness of the proposed method.

关 键 词：认知雷达 稳健波形 低峰均比 凸优化 半正定松弛 

分 类 号：TN958[电子电信—信号与信息处理]