A Hybrid Improved MVO and FNN for Identifying Collected Data Failure in Cluster Heads in WSN
Thi-Kien Dao, Jie Yu, Trong-The Nguyen, Truong-Giang Ngo
Abstract
This study proposes a new classifying approach for identifying collected failure data of cluster head (CH) in wireless sensor networks (WSN) based on hybridizing improved multi-verse optimizer (MVO) and feedforward neural network (FNN). An improvement of the MVO is proposed based on enhancing diversity agents for avoiding it's disadvanced of the local optimal. The data failure is detected for aggregating data in CH to forward to the base station (BS) based on classification by applying hybrid improved MVO and FNN. Twelve selected benchmark functions and the problem of identifying failure data in WSN are used in conducting comprehensive experiments to evaluate the performance of the proposed method. The experimental results are investigated and compared with the other approaches in the literature. The compared result exhibits the proposed technique that provides the alternative tool with the anticipation of influence on data sets and an effective way of forwarding the correct data from CH to BS in WSN applications.