Coalition Formation Game for Cooperative Cognitive Radio Using Gibbs Sampling

TL;DR

This paper models coalition formation in cognitive radio networks as a game and employs Gibbs Sampling to optimize secondary user assistance groups for maximizing data rates.

Contribution

It introduces a novel coalition formation game model for cognitive radio and applies Gibbs Sampling to efficiently find optimal coalition structures.

Findings

Gibbs Sampling effectively finds near-optimal coalitions.

Coalition formation improves secondary users' data rates.

The proposed method outperforms traditional approaches.

Abstract

This paper considers a cognitive radio network in which each secondary user selects a primary user to assist in order to get a chance of accessing the primary user channel. Thus, each group of secondary users assisting the same primary user forms a coaltion. Within each coalition, sequential relaying is employed, and a relay ordering algorithm is used to make use of the relays in an efficient manner. It is required then to find the optimal sets of secondary users assisting each primary user such that the sum of their rates is maximized. The problem is formulated as a coalition formation game, and a Gibbs Sampling based algorithm is used to find the optimal coalition structure.