Efficient modified-sinc filters for sigma-delta A/D converters
Letizia Lo Presti
Abstract
Letizia Lo Presti
Abstract
In this paper, a method to design and implement a very efficient multistage decimation filter for a sigma-delta (/spl Sigma//spl Delta/) A/D converter is proposed. The scheme is composed of two stages, but the proposed method can be easily extended to a multiple-stage implementation. The first-stage filter is obtained by properly rotating the zero-pole distribution of a comb filter in the z-plane. The obtained structure exhibits linear phase and can be implemented by using a recursive structure with only two multipliers. The design phase is easy and very flexible. As most of the quantization noise is eliminated at the first stage, the second-stage filter can be designed with relaxed specifications. Any classical design algorithm can be used for it. An alternative scheme for the second stage can be obtained by splitting the stage into two substages, and the method proposed in this paper can be iterated.
OpenAlex reports 149 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper, a method to design and implement a very efficient multistage decimation filter for a sigma-delta (/spl Sigma//spl Delta/) A/D converter is proposed. The scheme is composed of two stages, but the proposed method can be easily extended to a multiple-stage implementation. The first-stage filter is obtained by properly rotating the zero-pole distribution of a comb filter in the z-plane. The obtained structure exhibits linear phase and can be implemented by using a recursive structure with only two multipliers. The design phase is easy and very flexible. As most of the quantization noise is eliminated at the first stage, the second-stage filter can be designed with relaxed specifications. Any classical design algorithm can be used for it. An alternative scheme for the second stage can be obtained by splitting the stage into two substages, and the method proposed in this paper can be iterated.
Key concepts: Decimation, Delta-sigma modulation, Sinc function, Control theory (sociology), Quantization (signal processing), Filter (signal processing), Mathematics, Filter design