A delta-sigma-modulator feedforward network for a non-binary analog-to-digital converter

Abstract

A feedforward network, including first-order deltasigma modulators, is investigated to implement a non-binary analog-to-digital (A/D) converter. Weighting coefficients in the network are chosen to improve the conversion resolution, or the signal-to-noise ratio (SNR). Increasing a radix, as well as the number of network channels, improves the SNR. When the radix is three or more, the multi-levels after digitization are nonuniformly distributed. This can represent an analog signal more accurately than binary, uniformly distributed levels. Also, the effect of the characteristic mismatch in analog components on the SNR is analyzed. Our signal-level simulations confirm the analysis results.