Multi-objective distributed cooperative spectrum sensing for cognitive radio networks using greedy-based coalitional game

Abstract

In this work, we propose a distributed cooperative spectrum sensing (CSS) scheme for cognitive radio networks applying a greedy-based coalitional game. We formulate the cooperative spectrum sensing problem as a multi-objective optimization problem using the weighted global criteria method. The objective of the formulated problem is to jointly consider the relative impacts of the three key performance metrics in CSS: throughput gain, penalty cost, and energy overhead. By integrating these three distinct metrics into a global optimization problem, we aimed to achieve a well-balanced trade-off, essential for deploying the CSS scheme across diverse network and application environments. We prove that the formulated CSS problem is non-deterministic polynomially hard (NP-hard). The computational complexity of the formulated CSS problem is decreased by proposing a heuristic scheme that produces a sub-optimal solution in polynomial time. We develop the heuristic scheme by putting forward a greedy-based coalitional game theoretic model. The proposed scheme exploits the spatial diversity of secondary users (SUs) in forming distributed coalitions to minimize the effects of attenuation and distortion on the transmitted signal. Further sensing energy overhead is reduced by dynamically deciding the sensing duration for the SUs based on their received signal strength. To establish the stability of the coalitional game model, we carry out proof which shows that the proposed heuristic algorithm converges to a stable coalition structure. Simulation-based results show the effectiveness of the proposed scheme in terms of average detection accuracy, utility, and energy overhead compared to conventional schemes.