Research On Fast Mismatches Calibration Algorithm Of High Speed Sampling TIADC System

The bandwidth of next-generation aerospace,national defense,experiment and measurement systems is expanding from tens of MHz to instantaneous bandwidths in the hundreds of MHz and even GHz range.Development trends of phased array radar,5G wireless communication experimental systems,electronic warfare and digital oscilloscopes are pushing bandwidth to a higher position and significantly increasing the system's urgent need for high-speed,high-resolution analog-to-digital converters,making time-domain interleaved sampling become the focus of competitive research around the world.Analog to Digital Converter(ADC)is an important bridge between analog and digital signals,despite its development has been toward high-speed,high-precision and high-bandwidth direction,but due to the limitations of the existing process technology,the development of monolithic ADC will eventually reach the bottleneck,and for monolithic ADC,the resolution and conversion rate are always two mutually constraining indicators for monolithic ADCs.In order to meet the demand for high-speed and ultra-high-speed signal processing,multiple identical ADC time domain interleaved sampling(Time Interleaved ADC,TIADC)technology emerge as the times require.However,in actual production applications,the sampling rate of each sub-sampling channel of the TIADC can not achieve theoretical identical,which will lead into the sampling channel mismatch problem,and channel mismatch will cause serious performance impact on TIADC.Therefore,the compensation of channel mismatch has become a hot issue for domestic and international research.The thesis will conduct research revolves around a time-domain based blind correction adaptive algorithm for TIADC inter-channel mismatch,and focuses on the following work.1)Introducing the TIADC sampling principle and performing inter-channel mismatch error analysis.The theoretical modeling of the errors introduced by DC bias,gain and time mismatch is completed,and the characteristics of the errors introduced by the three mismatches and their effects on the TIADC sampling output signal spectrum are also analysed through Matlab simulations,while the correction schemes for DC bias and gain mismatch errors are introduced.2)Statistical-based time mismatch correction algorithm.The algorithms include first-order statistical-based and second-order statistical-based methods.The correction principles of the two methods are not only analysed in mathematical theory,but also validated by simulation in Matlab and compared with each other.3)An adaptive blind correction algorithm based on orthogonality is proposed.Compared with the first-order statistical and autocorrelation methods,the algorithm is based on orthogonality and shows excellent performance in the correction of full-band signals,especially high-frequency signals,while overcoming the high resource consumption of traditional second-order statistical methods.Secondly,the method is extremely scalable and can be modularly extended to multi-channel TIADC systems using the correction structure of a two-channel TIADC.(4)The implementability of the algorithm is verified.The proposed orthogonality-based adaptive blind calibration algorithm is verified to be realisable by using Verilog hardware description language for FPGA logic programming(Field Programmable Gate Array),Modelsim simulation and Matlab performance calculations on the host computer.