Capacitor exchange method for improving DNL (Differential Nonlinearity)/INL (Integral Nonlinearity) of successive approximation analog to digital converter

Abstract

The invention discloses a capacitor exchange method for improving the DNL (Differential Nonlinearity) / INL (Integral Nonlinearity) of a successive approximation analog to digital converter (ADC), and relates to the fields of microelectronics and solid-state electronics, in particular to the field of high-performance ADCs. The DNL / INL error caused by judgment of a secondary high order can be offset by exchange of a secondary high-order capacitor with all subsequent capacitors between two conversions without introduction of any extra ADC, introduction of any correction algorithm or splitting of any capacitor. Compared with a conventional correction method for improving the DNL / INL with dependence on an auxiliary ADC, a correction algorithm or capacitor splitting, the capacitor exchange method has the effects of simpler structure, smaller chip occupation area and higher easiness in on-chip implementation.

Description

technical field [0001] The invention relates to the field of microelectronics and solid state electronics, especially the field of high-performance analog-to-digital converters. Background technique [0002] The ADC converts the analog signal into a digital signal, which is a process of anti-aliasing filtering, sampling, holding and encoding. The analog signal passes through the anti-aliasing filter and the sample-and-hold circuit, and first becomes a ladder-shaped signal, and then passes through the encoder to convert each level in the ladder-shaped signal into a corresponding binary code. Its working principle is as follows: figure 1 As shown, among them, the anti-aliasing filter filters out the out-of-band signal in the input signal to prevent it from being mixed into the effective bandwidth, the sample-and-hold circuit samples the continuous signal to discretize the continuous signal, and the quantization circuit converts the discrete sample The signal is converted into...

AI Technical Summary

Problems solved by technology

The "minimum mean square error" algorithm generally requires hundreds of thousands of samples to converge, which has great limitations in practical applications

Images