Advances on Medium Access Control for Cognitive Radio Networks

Abstract

Over the last two decades, wireless communications have gained more and more importance in our everyday life. The rapidly growing of mobile wireless access services, together with an outdated fixed spectrum access policy is speeding up the need for more radio spectrum, a finite resource that is the basis of wireless communications. To increase the spectrum utilization, and therefore its usage efficiency, a big effort has been employed to develop and improve novel and alternative spectrum management policies, often referred to as Dynamic Spectrum Access (DSA). In DSA schemes, one piece of spectrum can be allocated to one or more users, which are called Primary Users (PUs) or licensed users, using the traditional fixed spectrum access policy. However, these users do not own the exclusivity of accessing the licensed channel, although they have higher priority in using it. But when the licensed channel is not being used by the PUs, the channel is vacant, leading to a spectrum underutilization. In this case, other users, denominated as Secondary Users (SUs) or unlicensed users, can dynamically access to the temporary vacant channel without interfering with PUs, hence improving the spectrum efficiency. The aforementioned characteristics of SUs, and therefore the peculiar characteristics of Cognitive Radio Networks (CRNs) distinguish them from the traditional wireless networks. Thus, traditional cross-layered Physical (PHY)-Medium Access Control (MAC) architectures become unsuitable for this new type of networks. In CRNs, the PHY layer should be responsible to performspectrum sensing tasks, enabling Cognitive Radio (CR) users to identify the spectrum holes, and also to reconfigure and optimize the transceiver based on the results acquired from the environment. Regarding the MAC sublayer, it should be capable to choose between the spectrum sensing and the spectrum access. The sensing/access coordination should be based on a tradeoff between sensing opportunities requirement and spectrum access requirements. This chapter reviews the most popular decentralized CR MAC protocols already proposed for CR networks. We start with a brief overview describing the basics of the CR principles. Then, we describe the main features that characterize the different CR MAC protocols, emphasizing their strengths and weaknesses. A new taxonomy that classifies the described protocols in several dimensions is also proposed. Finally, we point out open research problems regarding the design of distributed CR MAC protocols.