Real‑Time and Energy‑Efficient AI‑Driven Spectrum Allocation for 5G and 6G Networks Using Generalized and Lightweight Reinforcement Learning Models

Abhay Chaturvedi, Jayashri Jaywant Gajare, S. Sureshkumar, S. Muthuselvan, A. Swathi, Syed Zahidur Rashid

2025

Abstract

The fast-developing 5G and even toward the newer 6G era wireless communication networks require intelligent, dynamic and efficient spectrum allocation schemes. Existing traditional rule-based and static solutions do not satisfy the scalability, latency, and energy efficiency demands from dynamic heterogeneous networks. This study introduces a new paradigm for real-time and energy-efficient spectrum allocation based on lightweight reinforcement learning models. Existing approaches are limited by either simulation environments or extensive computational requirements; in contrast, our solution focuses on generalizability through varied network contexts and low-latency decisions, while being usable in the real world and continuously updated. It is built to cater to next-gen use cases like ultra-reliable low-latency communication (URLLC) applications, massive Internet of Things (IoT), and intelligent reflecting surfaces (IRS). We show the effectiveness of the proposed approach with extensive evaluations under realistic 5G/6G settings, where we achieve gains in spectrum efficiency, convergence stability, and operational energy savings.

Paper Citation

in Harvard Style

Chaturvedi A., Gajare J., Sureshkumar S., Muthuselvan S., Swathi A. and Rashid S. (2025). Real‑Time and Energy‑Efficient AI‑Driven Spectrum Allocation for 5G and 6G Networks Using Generalized and Lightweight Reinforcement Learning Models. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 636-642. DOI: 10.5220/0013887600004919

in Bibtex Style

@conference{icrdicct`2525,
author={Abhay Chaturvedi and Jayashri Gajare and S. Sureshkumar and S. Muthuselvan and A. Swathi and Syed Rashid},
title={Real‑Time and Energy‑Efficient AI‑Driven Spectrum Allocation for 5G and 6G Networks Using Generalized and Lightweight Reinforcement Learning Models},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={636-642},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013887600004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Real‑Time and Energy‑Efficient AI‑Driven Spectrum Allocation for 5G and 6G Networks Using Generalized and Lightweight Reinforcement Learning Models
SN - 978-989-758-777-1
AU - Chaturvedi A.
AU - Gajare J.
AU - Sureshkumar S.
AU - Muthuselvan S.
AU - Swathi A.
AU - Rashid S.
PY - 2025
SP - 636
EP - 642
DO - 10.5220/0013887600004919
PB - SciTePress