Degrees of cooperation in dynamic spectrum access for distributed cognitive radios

Abstract

The Federal Radio Act under Federal Communications Commission (FCC) allows predetermined users the right to transmit at a given frequency. Non-licensed users are regarded as "harmful interference" and not allowed to transmit in a certain frequency bands. As the demands for wireless communication become more and more pervasive, the wireless devices must find a way for the right to transmit at frequencies in the extremely limited radio band. However, there exist a large number of frequency bands that have considerable, and sometimes periodic, dormant time intervals. In the literature, those frequency bands refer to spectrum holes [1,2]. So there is a dilemma that on one hand the mobile users have no spectrum to transmit, while on the other hand some spectrums are not fully utilized. In order to cope with the dilemma, the FCC has recently investigated the efficient spectrum usage for cognitive radios, which is a novel paradigm that improves the spectrum utilization by allowing secondary networks (users) to borrow unused radio spectrum from primary licensed networks (users) or to share the spectrum with the primary networks (users). As an intelligent wireless communication system, cognitive radios are aware of the radio frequency environment, select the communication parameters (such as carrier frequency, bandwidth and transmission power) to optimize the spectrum usage and adapt the transmission and reception accordingly. Cognitive radios can bring a variety of benefits: for a regulator, cognitive radios can significantly increase in spectrum availability for new and existing applications. For a license holder, cognitive radios can reduce the complexity of frequency planning, facilitate the secondary spectrum market agreements, increase system capacity through access to more spectrum and avoid interference. For equipment manufacturers, cognitive radios can increase demands for wireless devices. Finally, for a user, cognitive radios can bring more capacity per user, enhance inter-operability and bandwidth-on-demand and provide ubiquitous mobility with a single user device across disparate spectrum access environments. The process for spectrum access is first to sense what the available spectrum is, then to get access to some of the available spectrum, next to use the available spectrum and finally to release the used spectrum. Significant research is necessary to investigate how to dynamically access the spectrum, which enables the opportunistic management of radio resources within a single access system or between different radio access systems. As a result, dynamic spectrum access can improve spectral efficiency, increased capacity and improve ease of access to the spectrum. In the literature, much work [8,9] has been done for dynamic spectrum access. In this chapter, we classify some of the dynamic spectrum access techniques for cognitive radios, according to the degrees of cooperation. The relations between distributed cognitive radios ranges from complete autonomy and non-cooperation, to full obeyance to the centralized controller. Specifically, we will discuss the following techniques for different degrees of cooperation: 1. Non-cooperative competition (Sect. 9.2) 2. Learning for better equilibria (Sect. 9.3) 3. Referee mediation (Sect. 9.4) 4. Threat and punishment from repeated interactions (Sect. 9.5) 5. Spectrum auction (Sect. 9.6) 6. Mutual benefits via bargaining (Sect. 9.7) 7. Contract using cooperative game (Sect. 9.8) 8. centralized scheme (Sect. 9.9) There are some tradeoffs for different types of approaches. For example, for noncooperative competition, the transceiver is simple but the performance can be inferior due to the extensive non-cooperation. On the other hand, the centralized scheme can achieve the optimal solution, but it is necessary for extensive measuring of channels and signaling to exchange channel information. Our goals are to investigate those different approaches with different degrees of cooperation, study in which network scenarios the approaches fit most and understand the underlying tradeoffs for the wireless cognitive network design. © 2007 Springer Science+Business Media, LLC.