Abstract Cognitive Radio (CR) improves the efficiency of spectrum utilization by allowing non- licensed users to utilize bands when not occupied by licensed users. In this thesis, we address several challenges currently limiting the wide use of cognitive radios. These challenges include identification of unoccupied bands, energy consumption and other technical challenges. Improved accuracy edge detection techniques are developed for CR to mitigate both noise and estimation error variance effects. Next, a reduced complexity Simplified DFT (SDFT) is proposed for use in CR. Then, a sub-Nyquist rate A to D converter is introduced to reduce energy consumption. Finally, a novel multi-resolution PSD estimation based on expectation-maximization algorithm is introduced that can obtain a more accurate PSD within a specified sensing time.

Abstract Cognitive Radio (CR) improves the efficiency of spectrum utilization by allowing non- licensed users to utilize bands when not occupied by licensed users. In this thesis, we address several challenges currently limiting the wide use of cognitive radios. These challenges include identification of unoccupied bands, energy consumption and other technical challenges. Improved accuracy edge detection techniques are developed for CR to mitigate both noise and estimation error variance effects. Next, a reduced complexity Simplified DFT (SDFT) is proposed for use in CR. Then, a sub-Nyquist rate A to D converter is introduced to reduce energy consumption. Finally, a novel multi-resolution PSD estimation based on expectation-maximization algorithm is introduced that can obtain a more accurate PSD within a specified sensing time.

The finite strip method (FSM) is a very efficient numerical method employed for performing the structural analysis of slender structures, such as cable-stayed bridges; the strip discretization of the model allows for the usage of a lower number of degrees of freedom, in comparison with the finite element method (FEM), while, as it will be discussed in the current research, the results obtained from both methods are in relatively good agreement. Moreover, to address the latest developments in the area of smart construction materials used for long-span bridges, the fiber reinforced polymer (FRP) composites were implemented for the bridge deck modeling, as part of a hybrid composite FRP cable-stayed bridge, and an extend laminate integrated finite strip method (LFSM) was applied for estimating the static structural performance of the hybrid composite FRP long-span cable-stayed bridge under several concentrated and uniformly distributed loadings. The free vibrations analysis was conducted for the Kap Shui Mun Cable-stayed Bridge model, and the natural frequencies were compared with the ones obtained from an FE model of the same bridge. One of the advantages of using the integrated finite strip method is that number of vibration modes, which can be included in the dynamic analysis when the effect of a sweeping sinus and a seismic loading are investigated when a conventional FE analysis would fail to converge. The outcomes of this research will set the stage for the hybrid long-span cable-stayed bridges modeling by the laminate integrated finite strip method (LFSM) which is more efficient and straightforward than the finite element analysis, for performing the static, free vibration, time domain, and frequency domain analyses.