← 返回论文列表 📄 下载原文 PDF  ISSCC 2026 · 32.4
ISSCC 2026Session 32 · LOW-POWER NOISE-SHAPING ADCSData Converters28nm CMOS

A 103.9dB-SFDR 83.8dB-SNDR 3MHz-BW Multi-Bit Quadratic-Exponential Noise-Coupled IDSM with High Tolerance to DAC Non-Linearity

⚡ 本页包含 AI 生成的分析内容,仅供参考

📋 论文概要

本文提出了一种二次指数噪声耦合的增量型Sigma-Delta调制器(IDSM),实现了高于4阶的噪声整形效果,同时具有对DAC非线性高容忍度和低噪声代价。通过复用DAC2和加法器,提高了有效过采样率和能效,在28nm CMOS工艺下实现了83.8dB SNDR和103.9dB SFDR。

💡 主要创新点

核心指标
83.8dB SNDR, 103.9dB SFDR, 3MHz BW, 794μW power, 179.6dB/181.4dB FoM
工艺节点
28nm CMOS
重要性
发表年份
ISSCC 2026

🏷 关键词

增量型Sigma-Delta调制器噪声耦合DAC非线性容忍度二次指数噪声整形

📄 原文摘要

Abstract This paper presents a quadratic-exponential noise-coupled (NC) IDSM to achieve a quantization noise shaping effect greater than 4th order with OSR 22, while having high tolerance to DAC non-linearity and a small noise penalty factor. The DAC2 and adder in NC are reused for a larger effective OSR and improve energy efficiency. Implemented in 28nm CMOS, the IDSM achieves an SNDR/SFDR of 83.8dB/103.9dB, with a 3MHz bandwidth, and 794μW, resulting in Schreier FoMSNDR/FoMDR of 179.6dB/181.4dB. Incremental ADCs are increasingly attracting interest for their Nyquist-like single-shot conversion, easy multiplexing, and simple decimation filter, making them ideal for highresolution IoT devices, sensor interfaces, and instrumentation [1-4]. To boost resolution and bandwidth with high efficiency, high-order loop filters and multi-bit quantizers are frequently utilized in IDSMs.(However, the non-uniform weighting of high-order IDSMs

👥 作者与机构

Zhensheng Li1,2, Biao Wang3, Mingqiang Guo1, Rui P. Martins1, Sai-Weng Sin1

University of Macau, Macau, China, 2UM Hetao IC Research Institute, Shenzhen, China, 3Formula Microelectronics, Shanghai, China

分类:Data Converters · 年份:ISSCC 2026