Research On Digital Channelization Technique For Wideband Satellite Communications

In the field of satellite communications, the classical transparent payload and theregenerative payload are evolving towards the ideal software defined payload, and thedigital channelized payload comes across the developing period as an intermediate. It isthought as the next generation payload. On a digital channelized satellite payload, thedigital channelization technology is at the heart of the satellite onboard switching, andthus it is expected to be prosperous in future applications of mobile satellite communica-tions.The dissertation focuses on the digital channelization technique for wideband satel-lite communications. And the main contributions of this dissertation are as follows:Firstly, a novel digital channelization approach and the corresponding digital chan-nelizer are proposed as a key contribution, removing the constraints like uniform-bandwidth switching, less tra?c channel, e.g. from existing digital channelization ap-proaches. Therefore, the proposed digital channelizer can be applied in not only uniform-bandwidth onboard switching, but also nonuniform-bandwidth onboard switching. Basedon the theory of perfect reconstruction modulated filter bank and filter's polyphase de-composition, the proposed novel digital channelizer implements sub-channel splitting,sub-channel switching and sub-channel combining so as to ?exibly realize wideband sig-nal's onboard switching with comparatively lower complexity.Secondly, a min-max optimization criterion is put forward for a digital channelizedsatellite communication system to optimize the configuration of frequency resource andpower resource. With the proposed criterion, much improvement in the utility of fre-quency and power can be gained for the satellite communication system. Meanwhile,this criterion can be easily and conveniently generalized to analyze the system capacityof a classical transparent system.Thirdly, an e?ective approach is proposed to design perfect reconstruction proto-type filter. In this approach, the design of perfect reconstruction prototype filter is mod-eled as a nonconvex nonlinear programming problem. To solve this problem, a meta-heuristic algorithm named variable neighborhood search algorithm is proposed so as toobtain low-ordered perfect reconstruction prototype filter. Then with the help of least mean square error optimization algorithm, high-ordered prototype filter can be designede?ectively. Compared with existing approach, the perfect reconstruction prototype fil-ter designed with the proposed approach bears the superiority in high stopband attenua-tion and extremely small reconstruction error, which is the key element in designing thenonuniform-bandwidth digital channelizer proposed in this dissertation.Finally, a very preliminary investigation is made to the application of compressedsensing-based sub-Nyquist sampling in digital channelized satellite communications.Here, great e?ort is done to emphasize the necessity of applying sub-Nyquist samplingin wideband satellite communication system. And then a conceive is made to implementdigital channlization via compressed sensing-based sub-Nyquist sampling.