ABSTRACT
CONTENTS 14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 14.2 Challenges and Related Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 14.3 Self-Coexistence among DSA-Enabled Smart Grid Networks . . . . . . . . . . . . . . . . . . . . . . . 354
14.3.1 Decision problem of SG networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354 14.3.2 Self-coexistence game analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
14.4 Self-Coexistence Using Multi-Stage Interaction Game . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 14.4.1 System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 14.4.2 Game Settings for Homogeneous Band-Based Self-Coexistence Problem . . 357 14.4.3 Perception-Based Learning Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 14.4.4 Game Settings for Heterogeneous Band Based Self-Coexistence Problem . . 358 14.4.5 Regret minimization model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359
14.5 Simulation Model and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 14.5.1 Self-Coexistence Strategy Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 14.5.2 Multi-Stage Learning Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
14.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362 14.7 Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
14.1 Introduction Wireless spectrum, the most valuable commodity of the current communication era, has been inefficiently utilized in different point of time and space. This sporadic use of licensed spectrum and overuse of unlicensed bands may not fulfill the data demands of ever-increasing future wireless applications and services, unless the “white spaces” (unused bands) are managed appropriately toward spectrum sharing and usage. A large portion of the spectrum has been allocated statically for TV, government, defenses, public safety etc., and these are used very intermittently, resulting in spectrum under-utilization. Additionally, this policy does not permit access to the unused spectrum resources to meet high network demand by which spectrum owners can generate additional revenue. The suboptimality of such static spectrum allocation policy has led the Federal Communication Commission (FCC) to propose a dynamic spectrum access (DSA) policy, which is expected to overcome the issues caused by static policy via adopting cognitive radio (CR) technology [1]. The CRs (a.k.a. secondary users) are intelligent radios that can perform periodic sensing to detect the absence of licensed or primary user (PU) in a band so it can access the chunk for data communication in an non-interfering manner. Once PU resumes the transmission in its allocated frequency, the secondary user (SU) must sense and switch to a different band promptly [2]. This way, the PUs and SUs can coexist in the same spectrum space, and SUs ensure that PU’s ongoing communication will not be disrupted.
