Date of Award
Electrical and Computer Engineering
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Providing some techniques to enhance the performance of spectrum sensing in cognitive radio systems while accounting for the cost and bandwidth limitations in practical scenarios is the main objective of this thesis. We focus on an essential element of cooperative spectrum sensing (CSS) which is the data fusion that combines the sensing results to make the final decision. Exploiting the advantage of the superior performance of the soft schemes and the low bandwidth of the hard schemes by incorporating them in cluster based CSS networks is achieved in two different ways. First, a soft-hard combination is employed to propose a hierarchical cluster based spectrum sensing algorithm. The proposed algorithm maximizes the detection performances while satisfying the probability of false alarm constraint. Simulation results of the proposed algorithm are presented and compared with existing algorithms over the Nakagami fading channel. Moreover, the results show that the proposed algorithm outperforms the existing algorithms. In the second part, a low complexity soft-hard combination scheme is suggested by utilizing both one-bit and two-bit schemes to balance between the required bandwidth and the detection performance by taking into account that different clusters undergo different conditions. The scheme allocates a reliability factor proportional to the detection rate to each cluster to combine the results at the Fusion center (FC) by extracting the results of the reliable clusters. Numerical results obtained have shown that a superior detection performance and a minimum overhead can be achieved simultaneously by combining one bit and two schemes at the intra-cluster level while assigning a reliability factor at the inter-cluster level.
Salout, Noor Hisham, "Combined Soft Hard Cooperative Spectrum Sensing in Cognitive Radio Networks" (2017). Electronic Theses and Dissertations. 7295.