An Improved Energy-Efficient Device-to-Device Communication in Overlaying Cellular Networks

This purpose of this paper is to design an energy efficient clustering protocol for device-device (D2D) in an overlay cellular networks. The protocol is also aimed at increasing the capacity of the cellular network. In order to achieve this, a clustering algorithm is proposed using a combination of Euclidean distance and the received signal to interference noise ratio for its design. These parameters are combined with Q-learning to define an energy efficient protocol for D2D communication. The protocol Clustering Algorithm for D2D communication using Reinforcement Learning (CADREL) will reduce energy consumption in D2D communication in a co-located antenna system. It also improves the allocation of resources necessary for efficient data transmission as well as reduce the amount of data transmissions by intelligently electing cluster heads (CH) so as to minimize data collisions and enhance the lifetime of the network. A simulation experiment was conducted in order to compare the protocol with other state of the art clustering protocol using energy efficiency and channel capacity as the metrics. From the simulations carried out, it was observed that the proposed algorithm outperforms the other protocols by 23% and 34% respectively.