无线传感网中低功耗的多声源定位与跟踪  

Low Power Multi-source Localization and Tracking in Wireless Sensor Networks

作　　者：冯玉武 胡国华[1] 洪蕾 Feng Yuwu;Hu Guohua;Hong Lei(Hefei University,Hefei 230601,China)

机构地区：[1]合肥学院,安徽合肥230601

出　　处：《廊坊师范学院学报（自然科学版）》2022年第2期40-44,49,共6页Journal of Langfang Normal University(Natural Science Edition)

基　　金：安徽省高校自然科学研究项目(KJ2019A0838);安徽省质量工程项目(2019jyxm0337)。

摘　　要：针对传统的无线传感网中多声源定位与跟踪功耗高、定位精度低以及适应环境能力弱等问题,提出了基于完全分布式结构的低功耗鲁棒性多声源定位方法。首先构造完全分布式结构的无线传感器网络,然后建立含混合噪声的声源信号模型。在此基础上采用EM算法的E步完成传感器检测信号总能量的分解,针对每个节点利用M估计完成声源位置的估计;利用鲁棒性均值一致性算法对每个节点所得的估计值进行融合以得声源位置的全局估计值,进一步提高定位精度;最后利用仿真实验验证了算法的有效性。Aiming at the problems of high power consumption, low positioning accuracy and weak ability to adapt to the environment in traditional wireless sensor networks, a robust multi-source localization method with low power consumption based on fully distributed structure is proposed in this paper. Firstly, a completely distributed wireless sensor network is constructed. Then, a sound source signal model with mixed noise is established. On this basis, E step of EM algorithm is used to complete the decomposition of the total energy of the sensor detection signal, and M estimation is used to complete the estimation of the sound source position for each node and the robust mean consistency algorithm is used to fuse the estimated value of each node to obtain the global estimate value of the sound source location, which further improves the positioning accuracy. Finally, a large number of simulation experiments verify the effectiveness of the proposed algorithm.

关 键 词：无线传感网 多声源 定位与跟踪 低功耗 

分 类 号：TN913.2[电子电信—通信与信息系统]