Non-uniform Sampling And Reconstruction Of Multiband Sparse Signals

With the development and the revolution of the wireless communications. Great changes have taken place in people’s lives. People’s daily life is becoming more and more rely on digital devices and wireless communications products. But with the scarcity of spectrum resources and inappropriate resource allocation structure, the spectral efficiency problem is increasingly outstanding. Cognitive radio technology is provided to solve this problem. Cognitive radio is an intelligent radio that can detects available channels in wireless spectrum, then accordingly changes its transmission or reception parameters to allow more concurrent wireless communications in a given spectrum band at one location. The first condition of Dynamic Allocation is that the authorized users in the channel can be detected quick and accurate. So the high speed sampling ADC is needed. Therefore, it is crucial to sampling signal rapid, reduces the computational complexity and reduce error performance of the sampling system.This thesis centres on and spreads out from multi-frequency sparses signal and uniform sampling and reconstruction algorithm. Research the Non-uniform sampling and reconstruction of multiband sparse signals. The major study contents and conclusions are as follows:1. Consider the class of multiband signals, whose frequency support resides within several continuous intervals, spread over a wide spectrum. Introduce a feasible approach for the periodic non-uniform sampling and reconstruction of multiband sparse spectrum signals and a differential signal perception model based on spectrum sensing.2. Introduce a feasible approach for choosing a sample pattern with a low condition number. Then give the detailed steps of signal reconstruction. Use the exhaustive search and SFS search method to optimize the performance of the sampling and reconstruction system. At last, give the simulation results, verify the feasibility of the proposed algorithm.3. Current signal spectrum information is usually unknown,in order to improve computational efficiency,we introduce an energy detection model for spectrum sensing in cognitive radio systems. Furthermore, we also simulate and estimate the error of both Music and Capon algorithms and compare the different time consuming under a variety of conditions. Simulations have demonstrated that the performance of Capon algorithm is much superior compared to Music algorithm when SNR is large enough. Especially in some large information signal case, Capon algorithm will much faster than Music algorithm.