Multiobjective optimization based on self‐organizing Particle Swarm Optimization algorithm for massive MIMO 5G wireless network

Summary The development of future fifth‐generation (5G) wireless networks (WNs) is an active research area worldwide. The 5G network grants significantly upgraded necessities contrasted with those in present wireless systems. Although massive MIMO (mMIMO) incorporation in WN empowers one to encounte...

Full description

Saved in:

Bibliographic Details
Published in:	International journal of communication systems Vol. 34; no. 4
Main Authors:	Purushothaman, Kesavalu Elumalai, Nagarajan, Velmurugan
Format:	Journal Article
Language:	English
Published:	Chichester Wiley Subscription Services, Inc 10-03-2021
Subjects:	Algorithms Decision making massive MIMO MIMO communication Multiple objective analysis multiple objective optimization objective functions Particle swarm optimization self‐organizing map spectral efficiency Wireless networks
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Summary The development of future fifth‐generation (5G) wireless networks (WNs) is an active research area worldwide. The 5G network grants significantly upgraded necessities contrasted with those in present wireless systems. Although massive MIMO (mMIMO) incorporation in WN empowers one to encounter 5G network technical necessities, it should handle different challenges to increase the performance. In this paper, a novel multiple objective self‐organizing particle swarm optimizer (SOMPSO) is used to solve multiple objective functions such as user data rate, energy efficiency, spectral efficiency, and average area rate of 5G WN with mMIMO. Furthermore, a fuzzy decision maker is utilized to select a solution vector for getting the best compromising result. Our experimental outputs demonstrate that this SOMPSO is an efficient and promising method to solve multiple objective problems in 5G networks. This paper used a multiple objective self‐organizing particle swarm optimizer to solve four conflicting objectives such area rate, energy efficiency, spectral efficiency, and average area rate. For maintaining the diversity of the solution, the neighborhood relations are constructed in the decision space by integrating SOM with the MOPSO algorithm. Also, it keeps an elite archive, and the members of the archive are used for leading the swarm to search the best nondominated solutions in a dynamic manner.
ISSN:	1074-5351 1099-1131
DOI:	10.1002/dac.4725