Study On Mismatches Calibration Technology Of High Speed TIADC System

ADC(Analog-to-digital converter)as a key component of the digital signal acquisition system,widely used in radar,communications and image processing.With the rapid development of electronic technology and computer technology,higher and higher requirements have been put forward in the sampling rate and accuracy of ADC chips.However,limited by the current level of technology,monolithic ADC is often difficult to meet the requirements.TIADC(Time-interleaved ADC)parallel sampling system,with its multi-channel parallel sampling structure characteristics,can multiply the sampling rate on the basis of existing technologies.Therefore,it has always been a research hotspot in the field of high-speed and high-accuracy analog-to-digital converters.However,as the number of sub-channel ADC chips increases,not only the complexity of the system design will be increased,but also higher and higher requirements will be put on the overall matching of the system.When the sub-channel ADC chip fails to meet the system matching requirements,the channel mismatch error will occur in the system,resulting in non-uniformity of the signal obtained by sampling,which seriously reduces the dynamic performance of the system.Therefore,efficient TIADC system calibration techniques need to be studied to compensate for the effects of channel mismatch errors.Firstly,according to the sampling principle of the TIADC system,a four-channel high-speed TIADC system with a total sampling rate of 2GSPS is modeled in Matlab.The effects of offset error,gain error,and timing skew error on system performance are studied and analyzed;Secondly,based on the different applications of TIADC system,an error calibration algorithm based on data statistics and an adaptive calibration algorithm based on LMS(Least mean square)are proposed.The first algorithm calculates the three error estimates based on the statistical characteristics of the output data of the sub-channels and calibrates them in turn.This algorithm has a simple calibration structure and is easy to implement in hardware.It can accurately compensate for errors.The second algorithm considers the overall impact of the three mismatch errors.By constructing a new non-integer periodic delay filter model,the entire period delay amount and the fractional delay amount can be compensated separately.The algorithm can feedback the error value in time,which is suitable for the system with high real-time requirements.Finally,the 4GSPS TIADC data acquisition system is introduced,and the calibration results of the two algorithms are compared and analyzed to summarize their applicable conditions.The channel mismatch error calibration algorithms proposed in this paper can significantly improve the dynamic performance of the TIADC system,which is of great significance for the research of high-speed data acquisition systems.