Title: Resource optimisation of downlink NOMA using Elman's recurrent neural network channel estimation and emperor penguin optimiser power allocation method

Authors: Shyam Gehlot; Swapnil Jain

Addresses: Department of Electrical and Electronics Engineering (EEE), Shri Vaishnav Vidyapeeth Vishwavidyalaya, Indore, Madhya Pradesh, India ' Department of Electrical and Electronics Engineering (EEE), Shri Vaishnav Vidyapeeth Vishwavidyalaya, Indore, Madhya Pradesh, India

Abstract: In NOMA 5G technologies, optimal resource allocation executes for maximising the system performances. Resource allocation such as channel estimation and power optimisation is imperative because obtained multipath signals at the receiver are deeply mixed and cannot be distinguished. In this work NOMA-based effective Channel Estimation (CE) algorithm named Elman-Recurrent Neural Network (Elman-RNN) has been presented. At the same time, a novel power allocation algorithm based on multi-objective Emperor Penguin Optimisation (EPO) for channel gain and throughput improvement is suggested. These works are implemented in MATLAB Tool. Here channel performance metrics like BER, outage probability, sum rate, channel gain, throughput, and Signal to Noise Ratio (SNR) are measured. Finally, results are compared with existing methods, from that we must know our work and methods are efficient.

Keywords: 5G; channel estimation; EPO; emperor power optimiser; E-RNN; Elman's recurrent neural network; NOMA; non-orthogonal multiple access; RNN; recurrent neural network.

DOI: 10.1504/IJWMC.2021.120028

International Journal of Wireless and Mobile Computing, 2021 Vol.21 No.2, pp.170 - 181

Received: 10 Nov 2020
Accepted: 28 Oct 2021
Published online: 04 Jan 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article