Abstract
The analog to digital conversion is an elementary part of the modern electronic systems. Almost all existing ADCs (Analog to Digital Converters) are based on the uniform sampling theory. It makes the signal acquisition time-invariant. Therefore, it can render a useless increase of the system activity, especially in the case of sporadic signals. Thus, an adaptive rate ADC which is based on the cross-level sampling is devised. It can adapt its conversion activity according to the input signal local variations. Therefore, it provides an intelligent signal acquisition which leads towards an efficient solution. The proposed ADC performance is studied for a speech acquisition. Results show a drastic reduction in the acquired number of samples and therefore promise a significant enhancement in the system power efficiency compared to the classical approach. A method to measure the proposed converter resolution is described. Moreover, its design flow is also presented.