一种基于隐马尔可夫模型的目标轨迹跟踪算法  

Partial CSI based affinity propagation dynamic clustering algorithm

作　　者：张汝云[1] 许明艳[1] 江涛[1] 

机构地区：[1]国家数字交换系统工程技术研究中心,郑州450002

出　　处：《计算机应用研究》2013年第5期1455-1457,1470,共4页Application Research of Computers

基　　金：国家"863"计划重点资助项目(2011AA010604)

摘　　要：现有的定位算法只表现了目标的静态位置特征,不能较好地在目标移动情况下跟踪获取目标的运行行为轨迹。针对这一问题,设计了一种基于隐马尔可夫模型的目标轨迹跟踪算法。该算法根据小区覆盖范围内的地理位置信息和目标用户的移动速度,建立隐马尔可夫模型;然后,基于维特比译码算法对最佳路径和最佳状态概率进行最优化求解,实现对目标轨迹的跟踪。仿真结果表明,该算法可以在基站小区分辨率下精确地获取目标轨迹。This paper designed a tracking algorithm based on hidden Markov model for mobile targets. Firstly, it established hidden Markov model according to the location information within cell covering area and target＇ s moving speed, then gained the optimization path and the optimization estate probability by employing Viterbi decoding algorithm, this method could track tar- get trace. The simulation results show that this algorithm can obtain accuracy target＇ s routing within base station cell resolving probability.

关 键 词：目标轨迹跟踪 隐马尔可夫模型 维特比译码 

分 类 号：TN911[电子电信—通信与信息系统] TP301.6[电子电信—信息与通信工程]