Cooperative Spectrum Prediction in Multi-PU Multi-SU Cognitive Radio Networks
Xiaoshuang Xing,Tao Jing,Wei Cheng,Yan Huo,Xiuzhen Cheng,Taieb Znati
DOI: https://doi.org/10.1007/s11036-014-0507-x
2014-01-01
Mobile Networks and Applications
Abstract:Spectrum sensing is considered as the cornerstone of cognitive radio networks (CRNs). However, sensing the wide-band spectrum results in delays and resource wasting. Spectrum prediction, also known as channel status prediction, has been proposed as a promising approach to overcome these shortcomings. Prediction of the channel occupancy, when feasible, provides adequate means for an SU to determine, with a high probability, when to evacuate a channel it currently occupies in anticipation of the PU’s return. Spectrum prediction has great potential to reduce interference with PU activities and significantly enhance spectral efficiency. In this paper, we propose a novel, coalitional game theory based approach to investigate cooperative spectrum prediction in multi-PU multi-SU CRNs. In this approach, cooperative groups, also referred to as coalitions, are formed through a proposed coalition formation algorithm. The novelty of this work, in comparison to existing cooperative sensing approaches, stems from its focus on the more challenging case of multi-PU CRNs and the use of an efficient coalition formation algorithm, centered on the concept of core, to ensure stability. Theoretical analysis is conducted on the upper bound of the coalition size and the stability of the formed coalition structure. A through simulation study is performed to assess the effectiveness of the proposed approach. The simulation results indicate that cooperative spectrum prediction leads to more accurate prediction decisions, in comparison with local spectrum prediction individually performed by SUs. To the best of our knowledge, this work is the first to use coalitional game theory to study cooperative spectrum prediction in CRNs, involving multiple PUs.