Cognitive Radios and Networks

In recent years, the Cognitive Radio and Cognitive Network paradigms have received significant attention by the research community. Cognitive Radios and Networks, in their initial formulation, are characterized by the addition of cognition capabilities such as reasoning and learning to wireless devices and networks, with the aim of providing enhanced adaptability and reconfigurability to cope with the ever-growing challenges of radio communications. The concepts of Cognitive Radio and Network have actually been interpreted in several different ways. In this thesis, we will first of all provide an overview of the different interpretations of Cognitive Radios and Networks, as appeared in the recent literature. We will then focus on the cognitive adaptation and reconfiguration of devices and networks by means of Artificial Intelligence (AI) techniques. In this respect, we will discuss how two well-known AI techniques, i.e., Fuzzy Logic and Neural Networks, can be used within a cross-layer and cross-device knowledge representation and reasoning architecture to become major enabling technologies for Cognitive Radios and Networks. For each technology we will discuss how it can be effectively adopted to implement key functionalities of cognitive systems, and we will present and discuss example applications such as cross-layer parameter optimization, wireless network access selection and channel assignment. For all the discussed applications, we will present performance evaluation results showing the advantages that the proposed techniques provide with respect to state-of-the-art approaches.