Background Compensation of Static TI-ADC Nonlinearities in Coherent Optical Receivers

This paper presents a novel background compensation technique of nonlinear distortions in time-interleaved analog-to-digital converters (TI-ADC). The proposed scheme is based on the well-known error backpropagation algorithm, which has been extensively used in machine learning applications. The technique enables adaptive compensation using low-complexity hardware such as look-up tables. The effectiveness and intrinsic background characteristics of this proposal make it a well suited candidate for application in high-speed, high modulation order digital communication receivers. Simulation results in a coherent optical receiver demonstrate that the proposed background compensation algorithm is able to practically eliminate the penalties caused by the TI-ADC nonlinearity.