Research And Implementation Of The Signal Receiving Method Based On Compressed Sampling

With the rapid development of information technology, the electromagnetic spectrum space is increasingly crowded. In order to meet the needs of more business, the frequency of radio frequency signal transmission is more and more high. In the field of cognitive radio and wideband reconnaissance receivers, the traditional Nyquist sampling system requires a very high sampling frequency for the high frequency transmission. And commercial analog to digital converter has been difficult to meet the requirements. To solve this problem, this paper designs and implements a compression sampling hardware system based on the theoretical framework of modulated wideband converters. The system can sample the sparse multi-band signal without distortion at a low sampling frequency and reconstruct the sub-band signals in the FPGA in real time. The system has practical significance for the practical application of compressed sampling technology.Firstly, this paper makes a survey of compressed sampling technique and chooses modulated wideband converter as research objects because the hardware implementation is more possible. After stuying the relevant principle of modulated wideband converter, this paper studies a variety of reconstruction algorithm under modulated wideband converter frame. Then the paper proposes an improved MMV-MOMP algorithm and chooses three kinds of reconstruction algorithms with high feasibility to make a performance simulation and comparative analysis. Meanwhile, the paper also makes a related research and simulation for the waveform recovery principle and carrier estimate algorithm based on compressed sampling.Then, this paper puts forward the overall hardware implementation scheme and key indicators and develops an 8-channel compressed sampling hardware system on the basis of theoretical research and simulation experiment. The system mainly comprises a CPCI chassis,8-channel high-speed random sequence generator module, 8-channel high-speed mixing module,8-channel baseband data processing module and motherboard and so on. This paper chooses the MMV-OMP algorithm as a real-time reconstruction algorithm of the hardware system. Based on Xilinx Xc6vlx475t FPGA chip, this paper presents the entire fixed-point implementation plan for back-end digital signal processing of the compressed sampling system, and gives details of the design for each module. In addition, the paper also developes the PC software which can restore the original sparse multi-band signal waveform and shows the waveform and spectrum.Lastly, this paper makes a detailed analysis and functional verification tests for each module to ensure the functional correctness of the compressed sampling circuit. Eventually, the paper completes the commissioning of the entire compressed sampling hardware system. By the way, the analog bandwidth of the sparse multi-band signal can be up to 2GHz and the original signal contains 3 sub-band signals. The system can show the time-varying waveform and spectrum of sparse multi-band signal on the PC Software. The processing bandwidth of the system is improved compared with the 1GHz processing bandwidth of the original system, and the system parameter configuration is more flexible.