Distributed Dynamic Adaptive-Neighbour Selection Strategy under Event-Trigger Communication

ShuWei Deng,Chao Zhang,Yao Mao
DOI: https://doi.org/10.1049/cmu2.12252
IF: 1.345
2021-01-01
IET Communications
Abstract:In wireless sensor networks, dynamic environmental changes may lead to the distribution of some sensors becoming very dense, which leads to the waste of communication resources in distributed estimation. In this paper, it is proposed that Event-triggered distributed dynamic adaptive-neighbour selection LMS (ET-dyANDLMS) algorithm can effectively solve the problems mentioned above. On the one hand, the network communication resources are reduced by the adaptive-neighbour selection. On the other hand, the network data redundancy is reduced by the event-trigger mechanism. From a theoretical point of view, the algorithm converges stably in mean meaning and mean square meaning, and the communication cost is reduced compared with other distributed algorithms. The theoretical analysis results verified by experimental simulation demonstrate the effectiveness of the proposed algorithm.
What problem does this paper attempt to address?