Power Control Based on Multi-Agent Deep Q Network for D2D Communication

TL;DR

This work addresses interference management in underlay D2D communications within a single cellular cell by casting power control as a multi-agent reinforcement learning problem. It adopts a Deep Q Network to replace explicit Q-tables, enabling scalable power optimization across D2D transmitters while enforcing cellular QoS via SINR thresholds. The method is evaluated against Open Loop and Max Power baselines, showing superior system and D2D throughput, especially as D2D density grows. The results highlight the potential of DRL-based power control to enhance spectrum efficiency and network performance in D2D-enabled cellular networks.

Abstract

In device-to-device (D2D) communication under a cell with resource sharing mode the spectrum resource utilization of the system will be improved. However, if the interference generated by the D2D user is not controlled, the performance of the entire system and the quality of service (QOS) of the cellular user may be degraded. Power control is important because it helps to reduce interference in the system. In this paper, we propose a reinforcement learning algorithm for adaptive power control that helps reduce interference to increase system throughput. Simulation results show the proposed algorithm has better performance than traditional algorithm in LTE (Long Term Evolution).

Figures (7)

• Figure 1: D2D communication multiplexing cellular system resources.
• Figure 2: D2D communication in cellular networks.
• Figure 3: System Model
• Figure 4: Reinforcement Learning model
• Figure 5: Deep Q Network