A/D converter background calibration algorithm based on neural network

With the development of A/D converter technology, the background calibration is a good way to enhance the circuit's performance. This paper presents a background calibration algorithm based on neural network to improve the ADC circuit's performance. The calibration algorithm can deal with the gain error, timing skew and comparator offset automatically. The simulation results show that the SNDR can be significantly enhanced through the background calibration algorithm. After completing the training of the calibration neural network, the algorithm can raise the SNDR from 55.16db to 73.11db and the ENOB from 8.58bit to 11.93bit.