相比相控阵雷达, 多输入多输出(MIMO)雷达发射功率密度低, 在低信噪比条件下空间分集特性优势不明显, 检测概率降低。针对此问题, 提出了基于逐目标消除-动态规划的多目标检测前跟踪(TBD)算法。该算法先利用统计量筛选航迹点, 对信号进行积累, 在降低TBD运算量的同时, 减小了噪声的积累; 再采用逐目标消除的思想, 克服了动态规划的航迹分叉和仅能处理单目标的缺陷, 实现了多目标的有效检测。文中还推导了目标检测的虚警概率和检测概率, 给出了检测门限与虚警概率的关系。复杂度分析和仿真表明, 相比传统TBD算法, 所提算法的运算量明显降低; 相比传统MIMO雷达检测方法, 在相同虚警概率和检测概率条件下, 所提算法要求的信噪比降低了5 dB, 有效提升了检测性能。

Compared with phased array radars, Multiple Input Multiple Output(MIMO) radar has lower transmit power density. In low SNR, the advantage of space diversity is not clear and the detection probability decreases. Aiming at this problem, a weak target Tracking Before Detecting(TBD) algorithm based on Successive Target Cancellation-Dynamic Programming(STC-DP) is proposed. Firstly, the algorithm screens the track points by statistics and integrates signals, reducing the complexity and the noise accumulation. Then, detect targets successively based on successive target cancellation. The proposed algorithm has advantages that it can overcome track furcation and detect multitarget effectively. The detection probability and false alarm probability are derived. The relationship between false alarm probability and detection threshold is given. Complexity analysis and simulation show the complexity decreases compared with traditional TBD algorithms; and the SNR is decreased by 5 dB in the same false alarm probability and detection probability compared with traditional detection algorithms, improving the detection performance.