Research On The Wideband Spectrum Sensing System Based On Modulated Wideband Converter

With the development of modern communication technology,radio technology had penetrated very deeply into all sections of society.The demand for high-speed wireless transmission technology is increasing.This makes the signal transmission bandwidth wider,and the frequency is more and more complex.Percept the current spectrum occupancy in real-time and then take access the wireless dynamically is to solve the current lack of an important means of spectrum resources.According to the traditional Nyquist sampling theorem,if there is no aliasing to get the spectrum within the broadband,you need a very high sampling rate and superior data processing capabilities.This is a huge challenge for existing analog-to-digital converters(ADCs),digital signal processors(DSPs),and memory.In recent years,compressed sensing technology develops rapidly in a variety of areas,and has achieved fruitful results.Sampling techniques based on compressed sensing can use a small amount of low-speed observation data to realize the sensing and reconstruction of wideband signals.The newly proposed Modulated Wideband Converter(MWC)has attracted much attention at home and abroad because of its simple and easy to implement structure and wide range of signal adaptability.At present,the research on MWC sampling technology mainly focuses on theoretical analysis,and there are still many problems in practical principle verification and prototype design.In this paper,the MWC sampling and reconstruction theory are studied in detail.Based on sparse Bayesian learning(SBL)reconstruction method is improved.The theory of channel compression and power spectrum reconstruction based on single channel Modulated Wideband Converter(SCMWC)are discussed.MWC prototype and SCMWC prototype design and implementation have also been achieved.The main research contents are summarized as follows:(1)Aiming at the noise folding phenomenon inherent in MWC system,the reconstruction method of SBL is introduced.Through the theoretical analysis of the phenomenon of noise folding,the conclusion of noise superposition comes from the process of compression.From the perspective of reconstruction,the detection probability is improved,and the SBL method is used to find the sparse solution.It is proved that the algorithm has better SNR adaptability through formula derivation and simulation.(2)According to the principle of compression sensing sampling,a set of low-rate multi-channel MWC principle verification system is designed.The system uses discrete component modules and experimental instruments built.The feasibility of MWC sampling principle in practical application is verified by the actual perception of the actual signal.(3)A prototype with 1GHz observation bandwidth based on SCMWC structure is designed.Then we gave a detailed introduction to each part of the design ideas.For the generation of periodic sequences and broadband mixing two design difficulties,select the appropriate chip to achieve the design indicators.Finally,the feasibility of the designed system is verified by experiments.