摘要
提出了一种适用于无线传感器网络中基于网格的目标跟踪算法,以解决在目标跟踪过程中信任度(belief)更新和传感器节点信息贡献量估计问题。该算法对信任度进行非参数化表示,用基于网格的算法对序列贝叶斯滤波过程进行实现。并且利用目标位置预测和基于网格的算法在不预先获知传感器节点测量数据的情况下,对节点的信息贡献量进行估算。在资源受限的无线传感器网络中,该算法在降低计算复杂度、提高算法适用范围方面都有显著改进。最后在仿真环境中验证了基于网格的目标跟踪算法的有效性。
In this paper , we presented a Grid-Based Algorithm for Target Tracking in Wireless Sensor Networks ,focusing on updating belief and estimating node's information contribution. Using a nonparametric representation for belief, we proposed a Grid-based algorithm for Sequential Bayesian filtering. Measure of information contribution can be estimated without having to first communicate the sensor data through Grid-Based algorithm and Target Movement Prediction Algorithm. The Grid-Based Algorithm admits an efficient computation to decrease complexities and generalize applicability. Finally, the simulation tests the performance of the Grid-Based algorithm.
出处
《南京邮电大学学报(自然科学版)》
2007年第6期1-6,共6页
Journal of Nanjing University of Posts and Telecommunications:Natural Science Edition
基金
国家高技术研究发展计划(863计划)(2002AA121068)资助项目