Clock Data Recovery Bias Control for Noise and Frequency Offset

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional clock data recovery devices are limited by the bandwidth of their low-pass filters, which restricts their ability to reduce noise and track larger frequency offsets, especially in severe environment conditions such as rapid temperature changes.

Innovation Solution

A clock data recovery apparatus comprising an oscillator, a sampler circuit, and a frequency control circuit that generates a bias voltage through integration calculation, digital-to-analog conversion, and low-pass filtering using two different driving paths with varying gains and low-pass filter bandwidths to generate a bias voltage, enabling wider tracking range and noise reduction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the low-pass filter is implemented with a wider bandwidth, then the clock data recovery device can track larger frequency offset, but it cannot reduce more noise

Engineering Contradiction:
Improvetracking rangeVSAvoidnoise
Core Design Contradiction:
Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent divides the frequency control circuit into two separate driving paths: a first driving path with a first low-pass filter having a first bandwidth, and a second driving path with a second low-pass filter having a second bandwidth. Each path processes the sampling signal independently with different filtering characteristics, allowing the system to simultaneously achieve noise reduction and frequency offset tracking capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different low-pass filter bandwidths to different parts of the system. The first low-pass filter uses a first bandwidth optimized for one aspect (noise reduction), while the second low-pass filter uses a second bandwidth optimized for another aspect (tracking capability). This local differentiation allows each path to excel at its specific function while working together to resolve the overall contradiction.

Inventive Principle:
Principle #3Local quality

2Object-affected harmful factors

If the low-pass filter is implemented with a narrower bandwidth, then the clock data recovery device can reduce more noise, but it cannot track larger frequency offset

Engineering Contradiction:
ImprovenoiseVSAvoidtracking range
Core Design Contradiction:
Object-affected harmful factorsVSAdaptability or versatility

Solution Approach 1:

The system segments the signal processing into two parallel paths with different low-pass filter bandwidths. The first path uses a narrower bandwidth for superior noise reduction, while the second path uses a wider bandwidth for better tracking of large frequency offsets. Both paths contribute to the overall system performance.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent merges the outputs of two driving paths with different low-pass filter characteristics into a unified frequency control mechanism. By combining the results from both paths, the system achieves both noise reduction capability (from the narrower bandwidth path) and frequency offset tracking capability (from the wider bandwidth path).

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS9923564B2Clock data recovery apparatus and method capable of reducing more noise as well as tracking larger frequency offsets
Publication Date: 2018.03.20 MEDIATEK INC
  • US9923564B2 patent drawing
  • US9923564B2 patent drawing
  • US9923564B2 patent drawing

AI summary

A clock data recovery apparatus includes an oscillator, a sampler circuit, and a frequency control circuit. The oscillator generates a clock signal according to a bias voltage. The sampler circuit samples an input data signal to generate a sampling signal according to the clock signal. The frequency control circuit generates the bias voltage by performing integration calculation, digital-to-analog conversion, and low-pass filtering for the sampling signal.