一种基于团簇中心点的时差量测滤波方法

Filtering Method for TDOA Measurement Based on Centre

在无源场景中,时差量测的时间间隔大多为非均匀特性。提出一种基于团簇中心点的时差滤波方法,将时间间隔过短的量测划为同一簇,每个簇的中心点代入滤波模型,实现了序贯迭代过程。相比于常用的滤波方法,避免了滤波发散问题,提高了时差的预测精度。最后,在仿真数据上将中心点滤波方法与卡尔曼滤波及改进算法进行比较,验证了该方法的准确性和有效性。

In the passive scene,the time difference of arrival(TDOA)measurements is mostly non-uniform in character.In this paper,a filtering method based on cluster centroids is proposed,where the measurements with short time intervals are classified as the same cluster,and the centroids of each cluster are substituted into the filtering model to realize the sequential iterative process.Our method,as opposed to the commonly used filtering methods,could avoid filtering dispersion and improve the prediction accuracy of time difference.Finally,the centroid filtering method is compared with Kalman filtering and the improved algorithm on simulation data to verify its accuracy and effectiveness.