摘要
为解决非高斯噪声背景下,基于贝叶斯Fisher信息矩阵和基于互信息的节点选择不一致的问题,该文提出一种基于多目标优化的节点选择方法。推导出节点噪声为混合高斯分布时的贝叶斯Fisher信息矩阵和互信息,将节点个数、选择的节点对应的Fisher信息矩阵和互信息共同作为优化的目标函数。提出利用基于分解的多目标优化方法寻找Pareto最优解,并采用与理想解相似的偏好排序技术(TOPSIS)从所有Pareto最优解中选择最终的节点选择方案。仿真实验结果表明,基于多目标优化的节点选择方法选择的节点具有更优更稳健的定位精度。
To overcome the flaw that the sensor selection methods based on either of Bayesian Fisher information matrix or mutual information could not provide coincident results,the multiple objective optimal technology is developed for sensor selection by minimizing the number of sensors,maximizing corresponding Bayesian Fisher information matrix and mutual information of the selected sensors.Then,the Technique for Order of Preference by Similarity to Ideal Solution(TOPSIS)approach is proposed to find the candidate that can better trade off the cost and two performance metrics.Comparison results demonstrate that the proposed method can find a better sensor group,and ultimately,its overall localization performance is more stable and accurate.
作者
闫青丽
陈建峰
YAN Qingli;CHEN Jianfeng(School of Computer Science and Technology,Xi’an University of Posts and Telecommunications,Xi’an 710121,China;School of Marine Science and Technology,Northwestern Polytechnical University,Xi’an 710072,China)
出处
《电子与信息学报》
EI
CSCD
北大核心
2021年第2期341-348,共8页
Journal of Electronics & Information Technology
基金
国家自然科学基金-浙江两化融合联合基金(U1609204)。
关键词
声源定位
到达方向角
节点选择
多目标优化
Acoustic source localization
Angle Of Arrival(AOA)
Sensor selection
Multi-objective optimization
作者简介
通信作者:闫青丽,女,1990年生,讲师,研究方向为信号与信息处理、无线传感器网络数据融合、目标定位与跟踪,gongchyy@163.com;陈建峰,男,1972年生,教授,研究方向为阵列信号处理、水下信号处理、声源定位、无线传感器网络信息融合等。
