Title: A multiple linear regression-based machine learning model for received signal strength prediction of multiband applications

Authors: M. Benisha; V. Thulasi Bai

Addresses: Research Centre – KCG College of Technology, Department of Information and Communication Engineering, Anna University, Chennai 600025, Tamil Nadu, India; Department of Electronics and Communication Engineering, Jeppiaar Institute of Technology, Tamil Nadu 631604, India ' KCG College of Technology, Karapakkam, Chennai, Tamil Nadu 600097, India

Abstract: In wireless communication, path loss prediction is of great impact to ensure service quality for users and performance optimisation. This requires a less complex and a more accurate path loss or received signal strength (RSS) prediction method. To deliver compliance, machine learning (ML) techniques have been considered. In this contribution, the principle behind ML-based RSS prediction and the procedure to correlate the antenna parameters well with the RSS value is presented for the designed multiband sub 6 GHz patch antenna, which can operate from 1 GHz to 6 GHz suitable for multiband applications. The regression-based ML method is used to train the model with simulated data and validated using Wi-Fi real-time RSS dataset. The same is extended for other frequency applications as well. From the predicted and measured values, it can be a best-suited model for the prediction of RSS thereby path loss for the future 5th generation wireless communications.

Keywords: path loss prediction; multiband antenna; machine learning; wireless communication; fifth generation (5G) mobile communication.

DOI: 10.1504/IJMC.2024.136626

International Journal of Mobile Communications, 2024 Vol.23 No.2, pp.127 - 147

Received: 02 Jun 2021
Accepted: 23 Mar 2022
Published online: 09 Feb 2024 *

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