基于改进粒子滤波的无线传感器网络目标跟踪算法
Target tracking algorithm for wireless sensor networks based on improved particle filter
  
DOI:
中文关键词:  粒子滤波;无线传感器网络;目标跟踪;集合卡尔曼滤波;人工鱼群
英文关键词:particle filter;wireless sensor networks;target tracking;ensemble Kalman filter;artificial fish swarm
基金项目:国家自然科学基金(61501107)资助项目
作者单位
邬春明 东北电力大学 信息工程学院,吉林吉林132012 
宫皓泉 东北电力大学 信息工程学院,吉林吉林132012 
王艳娇 东北电力大学 信息工程学院,吉林吉林132012 
赵星翰 国网吉林省电力有限公司,吉林长春130021 
郭立杰 吉林市供电公司,吉林吉林132001 
梁玉珠 吉林市供电公司,吉林吉林132001 
摘要点击次数: 1270
全文下载次数: 932
中文摘要:
      针对现有无线传感器网络目标跟踪算法中粒子滤波存在的粒子退化和样本贫化缺陷,提出一种改进的粒子滤波目标跟踪算法。通过采用集合卡尔曼方法获取目标状态的建议分布函数,利用集合的形式估计目标状态,同时结合最新观测数据不断修正分布函数,使跟踪精度大大提高;在重采样阶段引入人工鱼群算法优化粒子分布状况,使粒子更贴近真实值,增加了有效粒子数目,使粒子多样性得到增强,改善了粒子贫化问题。仿真结果表明,改进算法在跟踪精度、稳定性以及可靠性上均优于传统的目标跟踪算法。
英文摘要:
      Aiming at particle degradation and sample depletion defects in particle filter for existing wireless sensor network target tracking algorithm,an improved particle filter target tracking algorithm is proposed.By using the ensemble Kalman method to obtain the proposed distribution function of the target state,the target state is estimated by the form of the set and the distribution function is often corrected by combining with the latest observation data.Finally,the target tracking accuracy is improved.At the resampling stage,the artificial fish swarm algorithm is used to improve the distribution of the particles to make the particles be closer to the real value,increasing the numbers of effective particles and the particle diversity,meanwhile improving the phenomenon of particle depletion.The simulation results show that the improved algorithm is superior to the existing target tracking algorithms in tracking accuracy,stability and reliability.
查看全文  查看/发表评论  下载PDF阅读器

你是第3763443访问者
版权所有《南京邮电大学学报(自然科学版)》编辑部
Tel:86-25-85866913 E-mail:xb@njupt.edu.cn
技术支持:本系统由北京勤云科技发展有限公司设计