M. Shahkhir, Mozamir and Rohani, Abu Bakar and Wan Isni Soffiah, Wan Din and Zalili, Musa (2021) Improved GbLN-PSO Algorithm for indoor localization in wireless sensor network. Journal of Communications, 16 (6). pp. 1-8. ISSN 1796-2021. (Published)
|
Pdf
20210525015209231.pdf Download (1MB) | Preview |
Abstract
Localization is one of the important matters for Wireless Sensor Networks (WSN) because various applications are depending on exact sensor nodes position. The problem in localization is the gained low accuracy in estimation process. Thus, this research is intended to increase the accuracy by overcome the problem in the Global best Local Neighborhood Particle Swarm Optimization (GbLN-PSO) to gain high accuracy. To compass this problem, an Improved Global best Local Neighborhood Particle Swarm Optimization (IGbLN-PSO) algorithm has been proposed. In IGbLN-PSO algorithm, there are consists of two phases: Exploration phase and Exploitation phase. The neighbor particles population that scattered around the main particles, help in the searching process to estimate the node location more accurately and gained lesser computational time. Simulation results demonstrated that the proposed algorithm have competence result compared to PSO, GbLN-PSO and TLBO algorithms in terms of localization accuracy at 0.02%, 0.01% and 59.16%. Computational time result shows the proposed algorithm less computational time at 80.07%, 17.73% and 0.3% compared others.
Item Type: | Article |
---|---|
Uncontrolled Keywords: | PSO; GbLN-PSO, IGbLN-PSO; TLBO; localization error; computation time |
Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
Faculty/Division: | Institute of Postgraduate Studies Faculty of Computing |
Depositing User: | Miss. Ratna Wilis Haryati Mustapa |
Date Deposited: | 07 Jul 2021 09:52 |
Last Modified: | 07 Jul 2021 09:53 |
URI: | http://umpir.ump.edu.my/id/eprint/31628 |
Download Statistic: | View Download Statistics |
Actions (login required)
View Item |