Wireless Communication Signal Processing Methods Based On Compressed Sensing

The traditional signal processing method is established on the basis of the Nyquist Sampling theorem, which means that in order to recover the original signal exactly, the sampling rate must be at least twice as much as the original signal bandwidth. Nevertheless, with the fast development of wireless communication technologies, the demand for information is greatly increasing and the signal bandwidth becoming wider and wider, which makes the sampling, transmission, storage and processing of the traditional signal processing methods facing serious challenges. Therefore, how to get enough information while effectively reducing the sampling rate has become a research hotspot in wireless communication field.Compressed sensing theory makes the reconstruction of high-dimensional sparse signals from low-dimensional observation signals possible. Applying the compressed sensing theory to wireless communication systems can not only decrease the sampling rate, but also reduce the amount of information in signal processing, and eventually improve the performance of the communication systems. Hence, this thesis focuses on the applications of compressed sensing in wireless communication filed, including the CS-based signal reconstruction method, the CS-based channel shorten method and the CS-based broadband spectrum sensing method. The main contributions of this thesis are the following:1. For the signal reconstruction problem, in order to improve the efficiency of signal reconstruction, a new algorithm named one projection subspace pursuit is proposed. Firstly, the iterative process of subspace pursuit is divided into one correlation-maximum process and two projection processes. Secondly, the projection processes on the support set are diminished to reduce the complexity, and eventually improve the efficiency of signal reconstruction. Lastly, the analysis of the performance index of existing reconstruction algorithms shows its deficiencies, and a more reliable reference index is proposed.2. For the channel shorten problem, a sparse equalizer is designed based on the proposed semi-fusion greedy pursuit algorithm to reduce the computational complexity of the communication system whose complexity grows rapidly with the number of the nonzero taps of the channel shorten equalizer. Firstly, the channel shorten problem is transformed into the minimization problem of the number of the equalizer’s nonzero taps according to the minimum mean square error criterion. Then, the lower bound of the equalizer’s sparsity is obtained by predicting estimation process. Finally, the nonzero taps is achieved through the backup reconstruction and the support set extending processes.3. For the broadband spectrum sensing problem, a new cooperative spectrum sensing method based on compressed sensing is proposed to improve the sensing efficiency. Firstly, the sampling rate is further reduced under the influence of a band-pass filter group united with the compressed sampling process. Secondly, the second users only need to reconstruct partial spectrum signals according to the information communication between the fusion center and the second users, which makes the computational complexity of signal reconstruction process significantly decreased. Lastly, the reliability of the method is improved through the cooperation among multiple second users.