Research On System Error Processing Based On Time Interleaved Data Acquisition

At present,due to the limitations of electronic technology and external environment,the performance of analog-to-digital converters(ADC)has been greatly restricted.For this reason,some scholars have proposed a time-interleaved parallel sampling technology,which uses the method of combining multiple ADCs to interleave the same sampled signal.This system is called a time-interleaved parallel sampling system(Time-interleaved ADC,TIADC).The sampling rate is increased exponentially.At the same time,AXIe high-speed acquisition is used to make the sampling rate more efficient.But for the single-chip ADC in the TIADC system,due to the complexity of its manufacturing process,the interference of the external environment,and the degradation of the conversion accuracy caused by the long-term use of temperature drift,each sub-channel in the system will have a mismatch in the signal processing performance.The errors caused by these mismatches greatly reduce the sampling rate of TIADC.Therefore,it is particularly important to conduct an in-depth analysis of the error of the TIADC system and seek to reduce the error.Error correction is divided into two processes: error estimation and error correction.This article introduces the offset error,gain error,and time error estimation and correction algorithms in the interleaved sampling system in detail,and optimizes the algorithms.And self-designed the program control interface,display the data collected by AD9226 and AXIe on the program control interface,perform error correction on the MATLAB software platform,and the results obtained show that the original error has been significantly improved.Next,a specific introduction to the work carried out in this article:1.First study and analyze the overall sampling and signal reconstruction process of TIADC,then study the sampling clock of the system,analyze and summarize the main parameter performance of the sampling system,signal-to-noise ratio(SNR),signal-to-noise distortion ratio(SNDR),spurious-free dynamic range(SFDR),effective number of bits(ENOB),introduced three kinds of errors in the system: offset error,gain error,time error,and conducted in-depth research on the sources of the three errors,And finally analyzed and summarized the AXIe hardware platform.2.Using sine fitting method and mathematical statistics method to estimate and analyze the three kinds of errors,and compare the two algorithms.Through the use of FFT(Fast Fourier Transform)algorithm,mathematical statistics algorithm,subspace projection algorithm for error correction,and optimization and improvement of mathematical statistics algorithm,subspace projection algorithm.3.Through the synthesis of previously studied and improved algorithms.Use FFT,mathematical statistics,and subspace projection algorithm to correct the original data in MATLAB.Perform dynamic parameter test and analysis on the corrected data.SNR increased from 57.49 d B to 58.36 d B on average,and ENOB increased from 9.31 bit to 9.41 bit on average.The test results show that several algorithms have strong optimization effects.