Deep Reinforcement Learning-Enabled Dynamic Spectrum Coordination for Energy-Efficient Underwater Cognitive Acoustic-Optical Networks

View at publisher Summarise with VIRAT
Authors: Judy Simon, Nellore Kapileswar
Cited by: 0
References: 24

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The growing need of the high-capacity and energy-efficient underwater communication process has expedited the studies on the clever spectrum coordination mechanisms. The present research work shows a Hybrid Deep Reinforcement Learning and Game-Theoretic Spectrum Coordination Framework (DRL-GTSCF) of Underwater Cognitive Acoustic-Optical Networks (UCAONs). The proposed hybrid framework, in contrast to the conventional deep reinforcement learning (DRL)-based frameworks using only spectrum adaptation, integrates cooperative game theory in the DRL decision making process in order to determine jointly optimal and fair energyaware communication among distributed underwater nodes. The cognitive nodes are independent agents of DRL capable of learning the best transmission policies depending on the environmental parameters of pressure, salinity, and optical attenuation. Simultaneously, the game-theoretic coordination layer is used to stabilise and distribute the spectrum fairly by parametrically representing the inter-node interaction as a noncooperative game played by Nash equilibria to use the solutions as the refinement of DRL policy. The proposed DRL-GTSCF framework outperforms the DRL and heuristic approaches in terms of energy efficiency (up to 47%), end-to-end latency (reduced by 34 %), and spectrum usage (improved by 29 %) in simulated experiments in terms of hybrid acoustic-optical channel models for marine environment. Learning intelligent gaming fairness provides a more efficient solution to autonomous, adaptive, energy consumption underwater communications, and the design of future Internet of Underwater Things (IoUT) systems to monitor the ocean, defend against adversaries, and explore.

Fields of Study & Subjects

Underwater Vehicles and Communication Systems Cognitive Radio Networks and Spectrum Sensing Optical Wireless Communication Technologies

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
238–243
Publication date
21 January 2026
Language
EN