"ON THE SECRECY ANALYSIS OF COGNITIVE RADIO NETWORKS"

Abstract

The increasing number of connected devices represents a major challenge for broadband wireless networks that would require a paradigm shift towards the development of key enabling technologies for the fifthgeneration wireless networks. One of the key challenges towards realizing the next-generation wireless networks, however, is the scarcity of spectrum, owing to the unprecedented broadband penetration rate in recent years. Cognitive radio has emerged as a promising solution to the current spectrum crunch. Assuming a spectrum sharing scenario, the unlicensed users, also known as secondary users, opportunistically access the spectrum of primary (licensed) users under the constraint of not causing harmful interference to them. Similarly to traditional wireless networks, cognitive radio networks (CRNs) could be vulnerable to several attacks that could disrupt their operation. Eavesdropping attack is one of the security threats that can occur at the physical layer. Therein, unauthorized users try to overhear the communication between legitimate users. Since the SUs have to continuously adapt their transmit power to avoid causing harmful interference to the PUs, ensuring the security at the physical layer becomes a challenging task. Although several research works have investigated physical layer security (PLS) of wireless communication networks, secrecy analysis of CRNs is among the hottest research topics that are in their infancy. Therefore, in this thesis, the PLS of several cognitive radio-based wireless communication systems has been investigated and some main techniques such as friendly jammer, space diversity, and energy harvesting have been considered for security enhancement purposes. The first phase of work focused on investigating the impact of exploiting a multi-antenna relay to forward the message from a source to the intended destination. Indeed, secrecy metrics have been derived by considering a generalized fading model namely, Nakagami-m. The second phase of investigation consisted in performing secrecy analysis of different EH-based CRNs. Specifically, focus was placed on deriving closed-form and asymptotic expressions for the secrecy outage probability, based on which the impact of different key parameters of the network was investigated and new insights were gained. Finally, we focused our efforts on investigating the impact of a friendly jammer on the secrecy performance of CRNs. Indeed, our aim was to continuously send an artificial noise that could be added to the eavesdroppers’ signal and thus decrease his signal-to-noise ratio. However, given the power adaption constraint of secondary users, we were uncertain whether a friendly jammer would contribute to the enhancement of the secrecy performance of CRNs. Therefore, secrecy metrics were derived based on which we were able to conclude meaningful insights as to when a friendly jammer could improve the secrecy of a given communication system. Moreover, another main contribution consisted in deriving a new and generalized expression for the intercept probability representing communication between two nodes, through the aid of a relay performing decode-and-forward protocol, in the presence of two eavesdroppers at the first and second hop