Research On Frequency Measurement System Based On Optical Compressed Sensing

The frequency measurement technology in communication is significant.With the increase of frequency,the electronic frequency measurement schemes have revealed many shortcomings and insufficiency such as high complexity,low speed and non-anti-electromagnetic interference.With the development of microwave photonics,the scheme of optical frequency measurement has gradually attracted attention,and the results have gradually increased.At the same time,compressed sensing is widely used in various fields with its excellent sampling performance.Compressed sensing theory allows the sampling of ultra-wideband sparse signals with extremely low sampling rates,which brings essential method changes to radar and signal processing.The trend of communication becoming more and more broadband makes microwave photonics and compressed sensing have more and more potential applications.Therefore,we propose a frequency measurement scheme based on compressed sensing.(1)Design a single-channel photonic modulated wideband converter(MWC),which realizes the frequency measurement of ultra-wideband spectrum with a single Analog-to-Digital Converter(ADC)at a low sampling rate.(2)Through theoretical derivation and simulation analysis,this paper studies three widely used recovery algorithms,Orthogonal Matching Pursuit(OMP),Multiple Signal Classification(MUSIC),and Sparse Bayesian Learning(SBL)algorithm.The advantages and disadvantages of these algorithms and the reasons are analyzed.Finally,MUSIC algorithm is more suitable for single-channel MWC systems.The iterative characteristics of the SBL algorithm are also analyzed.(3)This paper draws the difference between the probability-based method and the matrix-based method by analyzing the periodic characteristics of MWC in the frequency domain.