TABLE OF CONTENTS This chapter discusses the expectation of a fully functional cognitive radio, including the cognitive decision-making process using case-based theory and genetic algorithms (GAs), to solve the multi-objective optimization problem posed by such a radio. The presentation has as its basis the cognitive engine developed at the Virginia Polytechnic Institute and State University (Virginia Tech) Center for Wireless Telecommunications (VT-CWT).
This chapter focuses on the intelligent cross-layer optimization of physical (PHY) and link (or medium access control, MAC) layers. The reader is encouraged to think beyond these two layers and consider how other layers, particularly network and transport, can be adapted and optimized by using the same techniques.
Section 7.2 defines optimization for a cognitive radio. A discussion of the cognitive radio as a mix of artificial intelligence (AI) and wireless communications follows in Section 7.3. Section 7.4 addresses the PHY and MAC layers, and considers which measurable radio settings and specifications ("knobs") and which radio and channel performance measures ("meters") fall into which layer. Section 7.5 introduces multi-objective decision-making (MODM) theory to analyze the radio's performance, and presents the analogy of GA to represent the methodology. In Section 7.6, the tiered algorithm structure of the cognition loop, based on modeling, action, feedback, and knowledge representation, is explored in detail.
Tailoring GA techniques for a cognitive radio, based on the simple GA discussed in these earlier sections, is addressed in...
