Coalition Formation Game for Delay Reduction in Instantly Decodable Network Coding Assisted D2D Communications

TL;DR

This paper proposes a distributed coalition formation game approach to reduce communication delay in multi-hop, IDNC-enabled D2D networks, extending previous single-hop solutions to more realistic network scenarios.

Contribution

It introduces a coalition formation game model and a distributed merge-and-split algorithm for delay reduction in decentralized multi-hop D2D networks using IDNC.

Findings

The proposed method outperforms existing approaches in reducing communication delay.

The coalition formation approach effectively manages decentralized multi-hop D2D networks.

Numerical results validate the efficiency of the distributed algorithm.

Abstract

Consider a wireless broadcast device-to-device (D2D) network wherein users' devices are interested in receiving some popular files. Each user's device possesses part of the content which is acquired in previous transmissions and cooperates with others to recover the missing packets by exchanging Instantly Decodable Network Coding (IDNC) packets. Recently, a distributed solution, relying on a non-cooperative game-theoretic formulation, has been proposed to reduce the communication time for fully connected D2D networks, i.e., single-hop D2D networks. In this paper, we develop a distributed game-theoretical solution to reduce the communication time for a more realistic scenario of a decentralized and partially connected, i.e., multi-hop, IDNC-enabled D2D network. The problem is modeled as a coalition game with cooperative-players wherein the payoff function is derived so that decreasing individual payoff results in the desired cooperative behavior. Given the intractability of the formulation, the coalition game is relaxed to a coalition formation game (CFG) involving the formation of disjoint coalitions. A distributed algorithm relying on merge-and-split rules is designed for solving the relaxed problem. The effectiveness of the proposed solution is validated through extensive numerical comparisons against existing methods in the literature.