Clock Recovery Apparatus Using Frequency Domain Correlation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Modern communication systems face challenges in accurately recovering clock frequency due to polarization effects, dispersion, and noise, particularly in optical transmission systems, where half-Baud Differential Group Delay (DGD) and states of polarization (SOP) can impair timing recovery accuracy.

Innovation Solution

A clock recovery apparatus employing Fourier transformation, correlation processing, and filtering to determine the clock frequency by comparing correlation values in the frequency domain, using digital Fourier transformers, auto-correlation, and cross-correlation of polarization signals, with optional weighting and filtering to enhance spectral components and mitigate distortions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If square-timing recovery (STR) with nonlinear operations is applied to recover clock frequency, then timing recovery can be performed in time domain, but timing recovery accuracy deteriorates under polarization effects like half-Baud DGD and certain states of polarization

Engineering Contradiction:
Improvetiming recovery operationVSAvoidtiming recovery accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent transforms the signal from the time domain to the frequency domain using Fast Fourier Transform (FFT). By moving to the frequency domain, the method exploits spectral characteristics that are less susceptible to polarization effects. The clock frequency is then identified through correlation in the frequency domain, where spectral components provide more robust timing information despite DGD and SOP variations.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces a correlation processor as an intermediary between the FFT output and the final clock frequency determination. This correlation processor compares spectral components across different frequency bins to identify the true clock frequency, acting as a mediator that filters out spurious spectral components caused by polarization effects while preserving accurate timing information.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If digital Fourier transformation and correlation processing are used to improve timing recovery accuracy, then robustness to polarization effects improves, but device complexity increases

Engineering Contradiction:
Improvetiming recovery accuracyVSAvoidsignal processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the signal processing into distinct segments: FFT processing for frequency domain transformation, correlation processing for identifying clock frequency, and filtering for enhancing spectral components. Each segment handles specific aspects of the problem, making the overall complex task more manageable and implementable through modular processing stages.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the representation parameters of the signal from time-domain samples to frequency-domain spectral components. This parameter transformation fundamentally alters how timing information is extracted, converting the problem from finding timing peaks in the time domain to identifying spectral peaks in the frequency domain, which are more robust to polarization effects.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If spectral filtering is applied to enhance clock tone components before Fourier transformation, then signal quality improves, but processing time and computational load increase

Engineering Contradiction:
Improvesignal qualityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies spectral filtering as a preliminary action before the FFT transformation. By pre-filtering the signal to enhance clock tone components and suppress noise and interference, the subsequent FFT and correlation operations work with cleaner data, reducing the computational burden and improving convergence speed. This preliminary purification of the signal pays off in reduced processing time despite the added filtering step.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2436139B1Clock recovery apparatus
Publication Date: 2020.02.12 HUAWEI TECH CO LTD
  • EP2436139B1 patent drawingFigure 1
  • EP2436139B1 patent drawingFigure 2
  • EP2436139B1 patent drawingFigure 3

AI summary

The invention relates to a clock recovery apparatus being configured to recover clock information from an input signal. The clock recovery apparatus comprises Fourier transforming means (201) being configured to transform the input signal 5 into a frequency domain signal upon the basis of a Fourier transform, correlating means (215) being configured to correlate the frequency domain signal to obtain a correlation value associated with a certain frequency, and clock recovery means (217) being configured to determine whether the certain frequency corresponds to a clock frequency in order to recover the clock information. 10