Clock Recovery via Eye Diagram Array Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing clock recovery methods are ineffective for short or noisy signals, particularly in four-level modulation schemes, leading to inaccurate data retrieval due to insufficient zero crossing averages and spurious noise, resulting in poor error rates.

Innovation Solution

A method and apparatus that generate an eye diagram representation from sampled digital signals, convert it into an array of data elements, perform multiple measurements on the array, and select a clock sample based on combined measurement outputs, including normalization, occupancy determination, convergence point correlation, and zero crossing analysis, to accurately recover the clock signal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If traditional zero crossing averaging methods are used for clock recovery, then the method is simple to implement, but the accuracy deteriorates in short or noisy signals due to insufficient zero crossing averages and spurious noise crossings

Engineering Contradiction:
Improveimplementation simplicityVSAvoidclock recovery accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent segments the clock recovery process into multiple independent measurement components (zero crossing positions, signal magnitude, signal slope, eye diagram occupancy) rather than relying on a single averaging method. Each measurement type processes the signal differently and contributes to the final clock phase determination, allowing the system to overcome the limitations of any single method in noisy or short signal conditions

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent merges multiple measurement outputs (zero crossing positions, magnitudes, slopes, and eye diagram occupancy values) into a combined metric that determines the final clock phase. This combination approach integrates the strengths of different measurement methods while compensating for their individual weaknesses, particularly in short or noisy signal scenarios where any single method would fail

Inventive Principle:
Principle #5Merging (Combining)

2Measurement precision

If training sequences are used to improve convergence in clock recovery, then the convergence accuracy is improved, but the method becomes impractical for short signals due to the additional signal length requirement

Engineering Contradiction:
Improveconvergence accuracyVSAvoidapplicability to short signals
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent performs preliminary actions by pre-calculating and storing eye diagram occupancy values and other signal characteristics during the signal acquisition phase. These pre-computed values are then directly used for clock phase determination without requiring additional training sequences or iterative convergence processes, enabling effective clock recovery from short signals

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a simplified model of the signal characteristics through eye diagram representation, which captures the essential timing and amplitude information without requiring the full original signal. This copied representation can be analyzed independently to determine clock phase, eliminating the need for lengthy training sequences while maintaining accuracy

Inventive Principle:
Principle #26Copying

3Measurement precision

If multiple measurements and processing steps are performed on eye diagram data, then the clock recovery accuracy is improved for noisy signals, but the computational complexity increases

Engineering Contradiction:
Improveclock recovery accuracy in noisy signalsVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies different processing qualities to different parts of the data structure. The eye diagram data is organized into bins representing specific time intervals, and only the relevant bins containing actual signal transitions are processed in detail. This localized processing approach maintains high accuracy for critical measurements while reducing unnecessary computational effort in regions with no signal activity

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent transforms the raw eye diagram data into multiple derived parameters (occupancy values, zero crossing positions, magnitudes, slopes) and processes these transformed parameters rather than the original data. This parameter transformation simplifies the subsequent analysis and reduces computational complexity while preserving the essential information needed for accurate clock recovery

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2490364B1Clock recovery apparatus and method thereof
Publication Date: 2016.01.13 THALES HOLDINGS UK PLC
  • EP2490364B1 patent drawingFigure 1
  • EP2490364B1 patent drawingFigure 2
  • EP2490364B1 patent drawingFigure 3

AI summary

The present invention relates to a method of recovering a clock signal from a digital signal in a receiver, the method comprising generating a representation of an eye diagram from a plurality of symbols of said digital signal, wherein said plurality of symbols are sampled at a sample rate such that said representation of said eye diagram is generated by a plurality of samples, converting said generated representation into an array of data elements, said array having a first set of data corresponding to said plurality of samples, and a second set of data representing a plurality of data bins, wherein each of said plurality of data bins includes a bin count, performing a plurality of measurements on said array of data elements and/or said eye diagram to obtain a plurality of measurement outputs corresponding to said plurality of samples, combining said plurality of measurement outputs at each sample, and selecting a sample as a clock sample based on results of said combination.