对运动高重频辐射源的三星时差跟踪算法

The Tracking Algorithm of Triple Satellites Using TDOA Based on Moving High Pulse Repetition Frequency Radiation Emitter#br#

  • 摘要: 本文针对三星对高脉冲重复频率(high pulse repetition frequency, HPRF)运动辐射源进行时差(time difference of arrival, TDOA)跟踪时出现时差模糊的问题,提出了一种基于多次观测滤波的时差跟踪解模糊算法。本文首先给出时差模糊的数学模型,其中包括分析模糊时差出现的原因、介绍时差窗和模糊时差数的计算方法和将模糊时差用高斯混合模型(Gaussian mixed model,GMM)逼近。之后利用三星时差定位得到所有模糊的滤波初始值,接着利用容积卡尔曼滤波(cubature Kalman filter , CKF)结合高斯和滤波(Gaussian sum, GS)算法在模糊时差观测下对辐射源进行跟踪,每次滤波之后删除权值较小的高斯成分来减小计算量,取最后一次滤波结果的加权平均作为最终输出。实验表明本文算法相较于其他解模糊方法可以更好地解时差模糊。

     

    Abstract: Aiming at the problem of time difference of arrival (TDOA) ambiguity when triple satellites were tracking moving radiation emitter with high pulse repetition frequency (HPRF), this paper proposed a tracking algorithm to resolve TDOA ambiguity based on multiple observation filtering. Firstly, this paper gave the mathematical model of the TDOA ambiguity, including the causes of TDOA ambiguity, the calculation method of the TDOA window and the number of ambiguous TDOAs. Then, ambiguous TDOAs were approximated by the Gaussian Mixed Model (GMM). After that, all ambiguous initial values of filtering were roughly obtained by the localization algorithm based on TDOA of triple satellites, and the cubature Kalman filter (CKF) was combined with the Gaussian sum (GS) filter algorithm to track the radiation emitter under the ambiguous TDOA observation. Finally, the Gaussian components with smaller weights were deleted to reduce the amount of calculation, and the weighted average of last filtering results was taken as the final output. Simulations prove that the proposed algorithm can resolve TDOA ambiguity more effectively than other methods.

     

/

返回文章
返回