Cooperative Spectrum Sensing with Censoring of Cognitive Radios and MRC-Based Fusion in Fading and Shadowing Channels

Introduction

There is huge demand of spectrum in this decade due to increase in internet traffic and other new services However, good part of the spectrum has already been licensed/ leased to different government, semi government and private organizations. Although, it can be noticed that the licensed and leased spectrums are not efficiently used and hence efficient spectrum allocation and utilization policies are required from spectrum traders and researchers.In order to deal with this conflict between spectrum scarcity and spectrum under-utilization, cognitive radio (CR) has been proposed as a revolutionary technology for the next generation of wireless communication networks (Mitola, 1999), (Haykin, 2005). In order to guarantee that the operation of the primary users (PUs) is not affected, the secondary users (SUs) need to sense the presence of active PUs: this process is referred to as spectrum sensing.

It is necessary to detect the presence of PUs accurately and quickly in order to find available unused spectrum, which is called spectrum holes. This is done by “Spectrum sensing” an important feature of CR technology. Accurate sensing of spectrum holes is a hard task because of the time-varying nature of wireless channels (Cabric, 2004), including fading and shadowing. Presence of multi-path fading or shadowing in the sensing (S) channel (S-channel) between a PU and a SU, may limit the successful detection of the PU by a single SU (Digham, 2003). The detection/sensing performance can be improved, by limiting the negative impact of fading, if different SUs are allowed to cooperate by sharing their local sensing information on the activity status of PUs: this is the essence of cooperative spectrum sensing (CSS) (Akyildiz, 2006), (Ghasemi, 2005), (Zhang, 2008). More precisely, CR systems allow the CR users¹ to sense the spectrum of PUs opportunistically without creating any intolerable interference to PUs. In many wireless applications, it is of great interest to check the presence and availability of an active communication link when the PU signal is unknown. In such scenarios, one appropriate choice consists in using an energy detector (ED) which measures the energy in the received waveform over a proper observation time window (Urkowitz, 1967), (Digham, 2003). Therefore, CSS using EDs improves the detection performance when all CR users sense the PU individually and send their sensing information via reporting (R) channels (R-channels) to a fusion center (FC). In CSS systems, the sensing information on the PUs's activity status sent by several CR users is combined at the FC to obtain a global decision. In general, the sensing information reported to the FC by several CR users can be combined in two different ways: through (i) soft combining (Teguig, 2012), (Sun, 2011), (Nallagonda, 2013a) or (ii) hard combining (Choudhari, 2012(a)), (Choudhari, 2012(b)), (Choudhari, 2013), (Nallagonda, 2011b), (Nallagonda, 2012). In this book chapter, we focus on soft combining of spectrum sensing decisions from several CR users when the S- and R-channels are affected by fading and shadowing. Specifically, we study the impact of Weibull fading as well as Lognormal shadowing in the R-channels and the benefits of censoring the CR users on the basis of quality of the R-channels.