Pipelined Phase Rotator Architecture for Low-Jitter 200-Gb/s CDR

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data center communication hardware faces challenges in reducing power consumption while meeting increasing bandwidth demands, particularly in high-speed wireline transceivers where stringent jitter requirements are difficult to meet with traditional Phase Rotators (PRs).

Innovation Solution

The implementation of pipelined phase rotators combined with twin PRs and Finite Impulse Response (FIR) filters in a Clock-and-Data Recovery (CDR) loop, which reduces power and area consumption while improving jitter performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional Phase Rotators are used in high-speed wireline transceivers, then bandwidth demand can be met, but power consumption increases and jitter requirements are difficult to meet

Engineering Contradiction:
ImprovebandwidthVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The phase rotator is divided into multiple pipeline stages, where each stage processes a portion of the phase rotation task. This segmentation allows the system to achieve high-speed operation (meeting bandwidth demands) while reducing the power consumption of each individual stage, as the computational load is distributed across multiple smaller units rather than concentrated in a single high-power component.

Inventive Principle:
Principle #1Segmentation

2Productivity

If traditional Phase Rotators are used in high-speed wireline transceivers, then bandwidth demand can be met, but jitter requirements are difficult to meet

Engineering Contradiction:
ImprovebandwidthVSAvoidjitter performance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

By segmenting the phase rotator into multiple pipeline stages, each stage can be optimized for specific jitter characteristics. The multi-stage architecture allows for better control and reduction of accumulated jitter compared to a single-stage design, thereby meeting stringent jitter requirements while maintaining high bandwidth capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a Clock-and-Data Recovery (CDR) loop that uses feedback mechanisms to monitor and adjust the phase rotation process. This feedback control helps maintain stable operation and reduces jitter by continuously correcting phase errors, enabling the system to meet both high bandwidth and stringent jitter requirements.

Inventive Principle:
Principle #23Feedback

3Use of energy by stationary object

If pipelined phase rotators are implemented, then power and area consumption are reduced, but device complexity increases

Engineering Contradiction:
Improvearea consumptionVSAvoiddevice complexity
Core Design Contradiction:
Use of energy by stationary objectVSDevice complexity

Solution Approach 1:

The phase rotator is segmented into multiple pipeline stages, each handling a portion of the phase rotation computation. This segmentation reduces the area consumption of each individual stage while distributing the overall complexity across multiple simpler units, achieving a balance between reduced area and manageable device complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4430751B1A multiple stage pipelined phase rotator and its application in a clock and data recovery loop
Publication Date: 2025.04.09 CIENA CORP
  • EP4430751B1 patent drawingFigure 1
  • EP4430751B1 patent drawingFigure 2
  • EP4430751B1 patent drawingFigure 3

AI summary

Aspects of the subject disclosure may include, for example, implementing a first stage including a first number of phase rotators in parallel generating respective clock phases off set by a fixed amount; a second stage including a second number of phase rotators receiving outputs from the first number of phase rotators of the first stage, the second stage outputting a first weighted sum of respective clock phases generated by the second number of phase rotators. The subject disclosure further includes the second number of phase rotators being less than the first number of phase rotators, and a total number of bits dedicated to phase selection being split across the first stage and the second stage. Other embodiments are disclosed.