论文部分内容阅读
天波超视距雷达(OTHR)目标跟踪面临着“三低”(低检测概率、低数据率、低测量精度)和“多路径”(多条传播路径)的严峻挑战,准确的传播模式辨识与精确的目标状态估计是改善跟踪能力的关键。针对上述问题,提出了一种基于马尔科夫蒙特卡洛吉布斯(MCMC-Gibbs)采样的OTHR联合状态估计与模式辨识算法,该算法通过MCMC-Gibbs采样求取当次迭代当前拍最优的关联矩阵,进而利用同时多量测滤波进行状态和协方差更新,最后引入联合估计与辨识风险函数寻求最优的模式辨识与状态估计结果。不同仿真参数下仿真结果表明该算法的有效性,同时该算法在径向距和方位角估计精度上均高于多路径概率数据关联算法(MPDA),但这是以计算量为代价的。
The sky-wave OTHR target tracking poses a serious challenge of “three low ” (low detection probability, low data rate, low measurement accuracy) and “multipath ” (multiple propagation paths) Spread pattern recognition and accurate target state estimation are the keys to improving tracking ability. In view of the above problems, this paper proposes an OTHR joint state estimation and pattern recognition algorithm based on MCMC-Gibbs sampling, which obtains the optimal current picture of the current iteration through MCMC-Gibbs sampling Then the state and covariance update are performed by using multiple simultaneous filtering. Finally, the joint estimation and identification risk function are introduced to find the optimal model identification and state estimation results. The simulation results under different simulation parameters show the effectiveness of the proposed algorithm. Meanwhile, the algorithm is superior to MPDA in both radial and azimuth estimation accuracy, but this is at the cost of computation.