Compensation of Converter System Based on Neural Network

The behavior of the analog-to-digital converter (ADC) strongly affects the global performance of the modern communication systems in terms of metrological uncertainty Due to errors in the manufacturing process, all ADCs in a time interleaved ADC system are not exactly identical. It means that gain and offset mismatch errors are introduced in the digital signal processing. These errors cause distortion in the sampled signal. In this paper we present a method based on T-model neural network for compensation of the mismatch errors. The compensation can be working while the ADCs are still running. The method is also adaptive to slow changes of the errors parameters. Combining with prediction of signal based on chaotic neural network, we can reach obvious improvement for ADC system and save a large amount of calculating time. This method has been validated with simulations of a time interleaved ADC system