DFE Tap Weight Optimization for FEC-Protected Links

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

High-speed communications systems face severe error propagation conditions, particularly with four-level pulse amplitude modulation (PAM-4) signals, due to decision feedback equalizer (DFE) tap weights greater than 0.5, which can transition error distribution from random to burst mode, limiting system performance and being sensitive to channel configurations and variations.

Innovation Solution

Adaptive updating of DFE tap weights based on predefined error propagation conditions, using iterative schemes and feedback mechanisms to maintain optimal weights and prevent error propagation, especially for 'borderline' links operating close to the bit error rate (BER) limit, and monitoring link conditions for dynamic adjustments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If DFE tap weights greater than 0.5 are used to improve equalization performance, then signal equalization quality is improved, but error propagation conditions occur transitioning from random to burst mode

Engineering Contradiction:
Improveequalization qualityVSAvoiderror propagation
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent implements dynamic adjustment of DFE tap weights through iterative updating schemes that adapt to channel conditions. The system transitions from static tap weight settings to dynamic optimization, where tap weights are continuously adjusted based on feedback from error propagation detection and channel state monitoring, resolving the contradiction between maintaining high equalization quality and preventing error propagation.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of tap weight values from fixed predetermined values to dynamically optimized values within the range of 0.5 to 1.0. By implementing iterative update schemes that adjust tap weights based on error propagation conditions and channel characteristics, the system optimizes the parameter to simultaneously achieve high equalization quality and prevent burst mode errors.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If DFE tap weights are limited to prevent error propagation, then error distribution remains random mode, but communication system performance is unnecessarily limited

Engineering Contradiction:
Improveerror distribution stabilityVSAvoidsystem performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements dynamic tap weight adjustment that allows weights to exceed the traditional 0.5 limit when channel conditions permit. The iterative updating scheme enables the system to dynamically determine optimal tap weights above 0.5 for borderline links, thereby improving system performance while maintaining error distribution stability through continuous monitoring and adjustment.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent expands the acceptable parameter range for tap weights from the traditional maximum of 0.5 to values between 0.5 and 1.0. By implementing iterative optimization that adapts tap weights to specific channel conditions, the system achieves higher performance for borderline links while preventing error propagation through intelligent parameter management.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If maximum DFE tap value is set based on input signal error limit, then error propagation is controlled, but the solution is sensitive to channel configuration variations

Engineering Contradiction:
Improveerror propagation controlVSAvoidchannel configuration sensitivity
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic adaptation to channel configuration variations through iterative tap weight updating. The system monitors channel state and error propagation conditions in real-time, adjusting tap weights to maintain optimal performance across different channel configurations. This dynamic approach eliminates sensitivity to channel variations by continuously adapting to the actual channel state.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent employs feedback mechanisms where error propagation detection and channel state information are fed back to the tap weight adjustment algorithm. This feedback loop enables the system to automatically adapt to different channel configurations and maintain optimal performance, resolving the sensitivity issue by using real-time information to guide tap weight optimization.

Inventive Principle:
Principle #23Feedback

4Reliability

If single DFE tap is used to mitigate burst mode errors, then error propagation is prevented, but equalization capability is reduced

Engineering Contradiction:
Improveerror propagation preventionVSAvoidequalization capability
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic optimization of multiple DFE tap weights through iterative updating schemes. Rather than using a single fixed tap, the system dynamically adjusts multiple tap weights to achieve both error propagation prevention and high equalization capability. The iterative process optimizes the combination of tap weights to simultaneously satisfy both requirements.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3474507B1Equalizer optimization for FEC-protected communication links
Publication Date: 2020.06.10 CISCO TECHNOLOGY INC
  • EP3474507B1 patent drawingFigure 1
  • EP3474507B1 patent drawingFigure 2
  • EP3474507B1 patent drawingFigure 3~5

AI summary

A method is disclosed for configuring equalization circuitry of a communication device. The method comprises determining, for a predefined first value of a tap weight of a decision feedback equalizer (DFE) of the equalization circuitry, whether a predefined error propagation condition occurs. The method further comprises iteratively updating the tap weight according to a predefined scheme, wherein each update of the tap weight occurs responsive to determining that the predefined error propagation condition occurs for a current value of the tap weight. The method further comprises ceasing the updating of the tap weight responsive to determining a difference between two adjacent values of the tap weight is less than a predefined resolution limit.