首页> 外文学位 >Design techniques for high performance CMOS flash analog-to-digital converters.
【24h】

Design techniques for high performance CMOS flash analog-to-digital converters.

机译:高性能CMOS闪存模数转换器的设计技术。

获取原文
获取原文并翻译 | 示例

摘要

This thesis describes research on very-high-sampling-rate, moderate-resolution, CMOS, analog-to-digital converters (ADCs). The sampling rate of an ADC can be extended through the use of a faster process technology (e.g., GaAs), or through time-interleaving. However, circuit techniques that extend the sampling rates of an ADC for a given CMOS technology, are still fundamentally required. Flash ADCs generally achieve the highest sampling rate. Architectures and building blocks of flash ADCs are investigated.; The comparator is one of the most critical components in a flash ADC. New circuit techniques that improve comparator performance are proposed. The addition of on-chip, compact, low Q inductors improves the sampling speed of the comparators without an increase in power consumption. Compact inductors consume 1% of the die area required for conventional on-chip inductors. The value of inductance is optimized through the consideration of both tracking and regenerative time constants. A clocked-cascode structure in the preamplifier reduces kickback to the reference. The use of a reduced swing sampling clock further extends sampling rate.; The use of minimum length transistors also helps to extend sampling speed and improve power efficiency. Offset correction is required even for moderate resolutions, because of the significant mismatch of minimum length transistors in deep-submicron CMOS. Comparator redundancy, DAC trimming at the comparator output, and DAC trimming at the reference input are proposed to calibrate comparator offset. Offset correction and calibration are optimized to maximize ADC yield and to minimize DNL and INL errors.; Two non-interleaved prototype flash ADCs were designed, fabricated and tested. A 4 bit ADC was implemented in 0.18 mum TSMC CMOS and a 5 bit ADC was implemented in 90 nm Intel CMOS. The 4 bit ADC achieves a sampling rate of up to 4 GHz with a measured effective resolution of 3.89 bits. The ADC consumes 551 mW at 3 GS/s with a 1.5 GHz full power input, and of this, the analog portion consumes 78 mW. The 5 bit ADC also achieves a sampling rate of up to 4 GHz and an effective resolution of 4.28 effective bits. This ADC consumes 227 mW at 3.5 GS/s with 1 GHz full power input. The analog circuitry of the 5 bit ADC consumes 115 mW. Both these ADCs achieve over twice the sampling rate of recently published state-of-the-art, non-interleaved, low-resolution CMQS ADCs. (Abstract shortened by UMI.)
机译:本文介绍了对超高采样率,中等分辨率的CMOS模数转换器(ADC)的研究。可以通过使用更快的处理技术(例如,GaAs)或通过时间交织来扩展ADC的采样率。但是,从根本上仍然需要用于给定CMOS技术扩展ADC采样率的电路技术。闪存ADC通常达到最高采样率。研究了闪存ADC的架构和构建块。比较器是闪存ADC中最关键的组件之一。提出了改善比较器性能的新电路技术。片上紧凑型低Q电感器的添加提高了比较器的采样速度,而不会增加功耗。紧凑型电感器消耗传统片上电感器所需的芯片面积的1%。通过同时考虑跟踪和再生时间常数来优化电感值。前置放大器中的时钟共源共栅结构可减少对参考的反冲。减少摆幅采样时钟的使用进一步扩展了采样率。使用最小长度的晶体管还有助于延长采样速度并提高功率效率。由于深亚微米CMOS中最小长度晶体管的严重不匹配,即使对于中等分辨率,也需要进行偏移校正。建议使用比较器冗余,比较器输出处的DAC调整和参考输入处的DAC调整来校准比较器失调。优化了失调校正和校准,以最大限度地提高ADC产量,并最大程度降低DNL和INL误差。设计,制造和测试了两个非交错原型Flash ADC。在0.18微米TSMC CMOS中实现了4位ADC,在90 nm Intel CMOS中实现了5位ADC。 4位ADC的最高采样率为4 GHz,有效分辨率为3.89位。在1.5 GHz全功率输入下,ADC在3 GS / s的功耗为551 mW,其中模拟部分的功耗为78 mW。 5位ADC还可实现高达4 GHz的采样率和4.28有效位的有效分辨率。该ADC在1 GHz全功率输入下在3.5 GS / s时消耗227 mW。 5位ADC的模拟电路消耗115 mW。这两个ADC的采样率均是最近发布的最新技术,非交错,低分辨率CMQS ADC的两倍。 (摘要由UMI缩短。)

著录项

  • 作者

    Park, Sunghyun.;

  • 作者单位

    University of Michigan.;

  • 授予单位 University of Michigan.;
  • 学科 Engineering Electronics and Electrical.
  • 学位 Ph.D.
  • 年度 2006
  • 页码 120 p.
  • 总页数 120
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 无线电电子学、电信技术;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号