Rajesh Kumar Sharma

• Cognitive radio is a novel approach for improving the utilization of radio spectrum. Spectrum sensing is one of the most important components of cognitive radio. This thesis concentrates on spectrum sensing and secure communication, allowing cognitive radio (CR) users to robustly identify unutilized spectrum and to use that spectrum securely, thus maximizing spectrum utilization without harmful interference to primary licensed users. This thesis proposes a new Correlation-Sum (CorrSum) method is proposed and its performance is compared with existing methods. Collaborative sensing is discussed as an effort for mitigating challenges of spectrum sensing arising due to channel fading. Different fusion methods as well as decision statistic combining methods are analyzed and compared. Since the multinode shadowing correlation in real environments has an impact on the performance of collaborative sensing, an experimental characterization of the shadowing correlation is performed. A tighter bound for the performance of correlation-based detection (CBD) methods is developed based on a signal with random correlation and Neyman-Pearson (NP) detection under the assumption of correlation distribution information (CDI). Additionally, a measurement campaign is presented where radio frequency (RF) spectra in many bands of interest are measured throughout a large sub-urban environment, generating realistic models for the random signal correlation. The measurement-based model indicates limits on performance gains possible with correlation-based detection. Finally, security issues in CR are discussed and a physical layer secure key generation method termed as reciprocal channel key generation (RCKG) is proposed. A new wideband indoor MIMO measurement campaign in the 2.51 GHz to 2.59 GHz band is presented, whose purpose is to study the number of available key bits in both line-of-sight (LOS) and non-line-of-sight (NLOS) environments.