Performance Analysis of Cognitive Radio Networks Using Cross-layer Design Approaches

This thesis studies the performance of both opportunistic spectrum access (OSA) and spectrum sharing (SS) cognitive radio networks (CRNs) using cross-layer design (CLD) approaches in order to provide reliable and optimum packet transmission at the medium access control (MAC) layer with quality of service (QoS) guarantees, to optimize the secondary user (SU) performance given the primary user (PU) protection and to realize the impact of imperfect spectrum sensing at the MAC layer. The reliable and optimum packet transmission at the MAC layer with QoS guarantees is accomplished by the combination of adaptive modulation with hybrid automatic repeat request (HARQ) protocol in OSA CRNs. The SU’s performance is maximized by formulating a convex optimization problem for the SU’s capacity over the power control (PoC) and spectrum sensing (SpSe) that is solved with an iterative subgradient method which results in optimal power allocation and sensing threshold selection. Finally, the impact of imperfect SpSe at the MAC layer is realized by modelling the Carrier Sensing Multiple Access with Collision Avoidance (CSMA/CA) and the SpSe as two state Markov processes and thus obtaining the joint steady state distribution that encompasses the parameters of both mechanisms.