Survey on the State-of-the-Art in Device-to-Device Communication: A Resource Allocation Perspective

TL;DR

This survey reviews the latest resource allocation algorithms for D2D communication in cellular networks, focusing on their design, performance, and challenges, highlighting categories like optimization, NOMA, game theory, and machine learning.

Contribution

It provides a comprehensive classification and analysis of existing resource allocation algorithms in D2D communication, including their operational modes and future research challenges.

Findings

Algorithms are categorized into four groups: optimization, NOMA, game theory, and machine learning.

The survey identifies key parameters and features influencing resource allocation performance.

Open challenges and future directions for research in D2D resource management are discussed.

Abstract

Device to Device (D2D) communication takes advantage of the proximity between the communicating devices in order to achieve efficient resource utilization, improved throughput and energy efficiency, simultaneous serviceability and reduced latency. One of the main characteristics of D2D communication is reuse of the frequency resource in order to improve spectral efficiency of the system. Nevertheless, frequency reuse introduces significantly high interference levels thus necessitating efficient resource allocation algorithms that can enable simultaneous communication sessions through effective channel and/or power allocation. This survey paper presents a comprehensive investigation of the state-of-the-art resource allocation algorithms in D2D communication underlaying cellular networks. The surveyed algorithms are evaluated based on heterogeneous parameters which constitute the elementary features of a resource allocation algorithm in D2D paradigm. Additionally, in order to familiarize the readers with the basic design of the surveyed resource allocation algorithms, brief description of the mode of operation of each algorithm is presented. The surveyed algorithms are divided into four categories based on their technical doctrine i.e., conventional optimization based, Non-Orthogonal-Multiple-Access (NOMA) based, game theory based and machine learning based techniques. Towards the end, several open challenges are remarked as the future research directions in resource allocation for D2D communication.