CDR 论文阅读

  • ALL-DIGITAL CLOCK AND DATA RECOVERY
    • 1 研究背景
    • 2 系统架构
      • Phase Detectors
        • XOR phase detector
        • Tri-State phase frequency detector
        • Hogge 's phase detector
        • Tri-wave Phase Detector
        • Half-rate linear PD
        • Alexander PD or Bang-Bang PD or Early-Late PD
        • THalf-rate BBPD
        • Multi-phase and multi-bit phase detectors
      • Reference-less, full-rate CDR architectures
        • Fully-digital CDR systems :An all-digital DCO-based CDR architecture
        • Reduced-rate CDR circuits
        • DLL-based CDR circuits
        • Phase-selector CDR circuits
        • Injection-locked CDR architectur
        • Burst-Mode gated-oscillator CDR architectures
        • Variable-interval 3x-oversampling CDR architecture
        • Jitter tracking CDR architectures
        • All-digital eye-tracking DR architecture
        • Blind Oversampling CDR
      • Jitters
    • 3 模型

ALL-DIGITAL CLOCK AND DATA RECOVERY

SYED IRFAN AHMED, 2010, Doctor of Philosophy ,Carleton University

1 研究背景


The contributions from this work are listed as follows:
• An understanding is developed using mathematical analysis and modeling as to
why the oversampling CDR architectures progressively perform better with an
increasing OSR in the presence of random, deterministic and asymmetric jitter.
• A unified mathematical analysis is presented that binds all PD architectures using
OSR as a parameter. A system-level behavioral simulation flow emerges that
incorporates the exact and generalized n-x oversampling (NXO) PD equations
derived in this work to simulate CDR performance in the presence of the given
input jitter.
• An understanding develops of how the oversampling CDR architectures morph all
the way from the BBPD architecture to the linear PD architecture and how the
mathematics seems to have little choice but to behave.
• A novel all-digital Data Recovery (DR) architecture is implemented that establishes the data detection window using phases derived from a multi-phase oscillator. This imparts performance robustness over Process, Voltage and Temperature
(PVT) corners while rendering the DR circuit rate-agile, adaptive and portable.
• Novel simulation methods are introduced to reduce the simulation time for transient jitter tolerance (JTOL) characterization. The idea is to exploit the maximum
slope of the jitter sinusoid as this puts the most stress on the CDR tracking performance.

2 系统架构


Phase Detectors

XOR phase detector

Tri-State phase frequency detector

Hogge 's phase detector

Tri-wave Phase Detector

There are three challenges that need to be resolved when using full rate PDs. The
first issue is that the oscillator running at high speed often provides reduced spectral purity
and tuning range. The second issue is the speed requirement imposed upon the flipflops.
The third issue for a HPD is the systematic skew for zero phase difference as proportional
pulses are wider by the CLK-Q delay of the flipflop. At high bit rates, the ripple owing to
the half cycle delay between the proportional and reference pulses is also untenable as the
ripple on a traditional oscillator can unlock the CDR circuit. A reduction of Kvco can alleviate the ripple by reducing the CP current but CDR bandwidth is reduced as well。

Half-rate linear PD

Alexander PD or Bang-Bang PD or Early-Late PD

THalf-rate BBPD

Multi-phase and multi-bit phase detectors

Reference-less, full-rate CDR architectures

Fully-digital CDR systems :An all-digital DCO-based CDR architecture

Reduced-rate CDR circuits

DLL-based CDR circuits

Phase-selector CDR circuits

Injection-locked CDR architectur

Burst-Mode gated-oscillator CDR architectures

Variable-interval 3x-oversampling CDR architecture

Jitter tracking CDR architectures

All-digital eye-tracking DR architecture

Blind Oversampling CDR

Jitters

3 模型

CDR 论文阅读 1相关推荐

  1. [论文阅读] (09)SP2019 HOLMES Real-time APT Detection through Correlation of Suspicious Information Flow

    <娜璋带你读论文>系列主要是督促自己阅读优秀论文及听取学术讲座,并分享给大家,希望您喜欢.由于作者的英文水平和学术能力不高,需要不断提升,所以还请大家批评指正,非常欢迎大家给我留言评论,学 ...

  2. 论文阅读:Deep Learning in Mobile and Wireless Networking:A Survey

    论文阅读:Deep Learning in Mobile and Wireless Networking:A Survey 从背景介绍到未来挑战,一文综述移动和无线网络深度学习研究 近来移动通信和 5 ...

  3. WSDM 2022 推荐系统论文阅读(中)

    继续上一篇文章来阅读WSDM 2022中推荐系统的论文: WSDM 2022 推荐系统论文阅读(上) 声明:本论文阅读笔记主要是对论文摘要的概括,评论仅代表个人的阅读观点,欢迎大家在评论区讨论! 1. ...

  4. 论文阅读工具ReadPaper

    对于搞科研的同学们来说,看论文是要经历的第一关,尤其是要读好多篇论文的时候,更是着实令人头大. 这不,最近无意中发现了个在线论文阅读网站:readpaper.com,号称「论文阅读笔记神器,硕博科研学 ...

  5. 多目标跟踪:CVPR2019论文阅读

    多目标跟踪:CVPR2019论文阅读 Robust Multi-Modality Multi-Object Tracking 论文链接:https://arxiv.org/abs/1909.03850 ...

  6. 快速人体姿态估计:CVPR2019论文阅读

    快速人体姿态估计:CVPR2019论文阅读 Fast Human Pose Estimation 论文链接: http://openaccess.thecvf.com/content_CVPR_201 ...

  7. Action4D:人群和杂物中的在线动作识别:CVPR209论文阅读

    Action4D:人群和杂物中的在线动作识别:CVPR209论文阅读 Action4D: Online Action Recognition in the Crowd and Clutter 论文链接 ...

  8. 深度学习点云语义分割:CVPR2019论文阅读

    深度学习点云语义分割:CVPR2019论文阅读 Point Cloud Oversegmentation with Graph-Structured Deep Metric Learning 摘要 本 ...

  9. 3D目标检测论文阅读多角度解析

    3D目标检测论文阅读多角度解析 一.前言 CNN(convolutional neural network)在目标检测中大放异彩,R-CNN系列,YOLO,SSD各类优秀的方法层出不穷在2D图像的目标 ...

最新文章

  1. python字符照片_python图片转字符图片
  2. 成就更卓越、更有意义的人生
  3. CMake常见变量——Project和CMake相关信息
  4. WINCE6.0 Quarter VGA(QVGA) Resources组件
  5. 江苏省二级python考试题库_python二级考试试题.doc
  6. 1688 复杂业务场景下的 Serverless 提效实践
  7. Qt 信号量 QSemaphore Class 的翻译
  8. protobuf入门教程(六):导入定义(import)
  9. 线程安全-常用的模式
  10. 云服务器怎么执行sql文件在哪里,总结帝国CMS下在PHP文件中怎么调用数据库类执行SQL语句实例...
  11. Linux内存管理:MMU那些事儿
  12. 如何实现快捷方式中的查找目标功能
  13. 数据之路 - Python爬虫 - urllib库
  14. js中的null VS undefined
  15. oracle 中的除法函数,Oracle 函数
  16. ASO|几个ASO优化方案解析(下)
  17. 李白的苏台览古译文赏析
  18. 杭州电子科技大学ACM-1096
  19. python调用gephi_利用Python和Gephi制作人物关系网络图
  20. 扦插机器人_如何看待我国农业采摘机器人发展趋势?

热门文章

  1. 信号转换为图像 时域频域分析
  2. (线性代数笔记)2.阶梯型矩阵
  3. 数据结构c语言版课设,数据结构课程设计c语言版
  4. 精细化管理计算机室取得的成效,班级精细化管理的实施与成效
  5. 四十岁软件测试员技术支持,【软件测试工程师(成都)财务专员面试题目|面试经验】-看准网...
  6. oracle 延迟加载,延迟约束禁止直接路径插入
  7. Rufus 中文绿色版 - 简单快速制作 USB 启动盘软件 (U盘安装 Windows / Linux 系统)
  8. pioneer软件VoLTE测试步骤,VoLTE方案及测试分析指导(全面)
  9. python for语句 不带冒号_为什么Python语句块要有一个分号(:)?
  10. 英方软件:Data+引领数据复制的下一个十年