Research On The Key Technology Of Optoelectronic Analog-to-digital Conversion

Analog-to-digital conversion technology has become an important piece of the modern signal processing technology puzzle.And the role of optoelectronic analog-todigital conversion technology is increasingly prominent in high-speed sampling and digitization of broadband signals.All-optical quantization as a difficult point in optoelectronic analog-to-digital conversion has received a lot of attention from researchers.And for broadband systems,the introduction of undersampling technique can narrowband the broadband signal.Therefore,this paper focuses on two key technologies in optoelectronic analog-to-digital conversion,namely,all-optical quantization and undersampling,and conducts theoretical and simulation studies.The main research contents are as follows.Firstly,the theoretical basis of the optoelectronic analog-to-digital conversion is derived from the equations,and the performance index of the analog-to-digital conversion is discussed.The optical domain sampling principle required for the design of all-optical analog-to-digital conversion,the operating characteristics of optoelectronic devices,and the nonlinear effects in highly nonlinear optical fibers are also described.Secondly,a multi-threshold nonlinear all-optical quantization scheme based on phase coding is proposed for how to effectively enhance the quantization bits of the optoelectronic analog-to-digital conversion.This scheme achieves phase encoding and optical sampling by adjusting the bias voltage of multiple lithium niobate Mach-Zendel modulators.The sampled optical pulse peak power information from the modulator output is then mapped to frequency shift information by the soliton self-frequency shift effect in the highly nonlinear fiber.The peak power of the optical pulse is approximately linearly related to its frequency shift.Finally,the multi-threshold optical quantization is completed by the wave division device with the multi-threshold coding table calculated by the threshold distribution algorithm.This scheme enhances the number of quantization bits of the system by increasing the number of parallel channels and the number of threshold points.In the multi parallel channel system,the number of threshold points is increased to increase the quantization bit enhancement efficiency.log2(2MN)is achieved for the multi-threshold all-optical analog-to-digital conversion system with M channels and N thresholds.The upper limit of the number of threshold points is proven by simulation to depend on the material,structure and length of the high nonlinearity fiber used and the peak power of the mode-locked pulsed laser,and the simulation results show that up to 37 threshold points can be selected.When 37 threshold points are selected,only2 parallel channels can be used to achieve an effective bit count of 6.32 bits in the system.Due to the development of mode-locked pulsed laser technology,femtosecond pulsed light sources are widely used in all-optical quantization,but there is a problem of low repetition frequency,which makes such quantization schemes not suitable for broadband systems.Therefore,this paper investigates another key technique in analogto-digital conversion,namely,undersampling technique.Combined with the compressive sensing theory,a photonic undersampling system is proposed to estimate the spectrum of the original signal under undersampling conditions,solving the problem of insufficient sampling rate of femtosecond pulsed light sources.The simulation results show that the spectrum of the original signal recovered from the undersampled data is mainly affected by three factors: the sparsity of the original signal in the frequency domain,the signal length entering the photonic undersampling system,the number of observations in the photonic undersampling system.Finally,the constraint relationship between the three factors is analyzed in terms of the probability of successful signal reconstruction,and a fitting formula is given to ensure the success of signal reconstruction.