ABSTRACT
CONTENTS 13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 13.2 Game Theory and Distributed Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
13.2.1 Motivation of Using Game Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 13.2.2 Preliminary of Game Theory and Distributed Learning . . . . . . . . . . . . . . . . . . . . . 321
13.2.2.1 Basic Game Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 13.2.2.2 Distributed Learning for Achieving Equilibria . . . . . . . . . . . . . . . 322
13.3 Graphical Game for Spatial Opportunistic Spectrum Sharing . . . . . . . . . . . . . . . . . . . . . . . 323 13.3.1 System Model and Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 13.3.2 Local Altruistic Game for Network Throughput Maximization . . . . . . . . . . . . . 326 13.3.3 Local Congestion Game for Network Collision Minimization . . . . . . . . . . . . . . 326 13.3.4 Spatial Adaptive Play for Achieving Global Optimization . . . . . . . . . . . . . . . . . . 327
13.3.4.1 Convergence and Optimality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 13.3.4.2 Implementation Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
13.3.5 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 13.3.5.1 Convergence of the SAP-Based Learning Algorithm . . . . . . . . . 329 13.3.5.2 Throughput Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
13.3.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 13.4 Robust Game for Dynamic Opportunistic Spectrum Sharing . . . . . . . . . . . . . . . . . . . . . . . . 333
13.4.1 System Model and Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 13.4.2 Robust Game Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 13.4.3 Stochastic Learning for Achieving Nash Equilibria . . . . . . . . . . . . . . . . . . . . . . . . 336 13.4.4 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
13.4.4.1 Convergence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338 13.4.4.2 Throughput Performance of Homogeneous OSS Systems . . . . 339 13.4.4.3 Throughput Performance of Heterogeneous OSA Systems . . . . 340
13.4.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341 13.5 Future Directions and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
13.1 Introduction With the explosive increase in wireless transmission demands, traditional static, and pre-determined spectrum allocation approaches can not meet the requirements of flexibility, adaptability and intelligence for the terminals (users). Cognitive radio has been established as an innovative framework for intelligent opportunistic spectrum sharing (OSS) which could observe the environment, make intelligent decisions and reconfigure the hardware [1]. In OSS systems, the devices are required to be autonomous and smart. Basically, mutual interactions in the wireless environment, mainly including competition, interference and coordination, should be well addressed when all the devices are autonomous and smart. Game theory [2] is a powerful tool to study the interactions among multiple autonomous decision-makers and has been extensively applied in wireless communication networks. This chapter presents novel game models and discusses distributed learning techniques for OSS systems.
