Adaptive Equalizer Constellation Map Switching for Convergence Speed

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In adaptive equalizers, convergence speed adjustment and convergence error adjustment restrict each other, and the blind equalization convergence capability is not improved in prior art.

Innovation Solution

An adaptive equalizer that filters input signals using multiple filter coefficients and constellation maps, where the constellation map is adjusted based on error signals and buffer signals to optimize filter coefficients, maintaining inherent filter features while enhancing convergence speed and blind equalization convergence.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If the iterative update step parameter or iterative update algorithm of the coefficient updating unit is modified to improve convergence speed, then convergence speed is improved, but convergence error performance deteriorates

Engineering Contradiction:
Improveconvergence speedVSAvoidconvergence error
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent segments the equalization process into two distinct phases: a blind equalization phase using a simplified coefficient updating unit without decision feedback, and a decision-directed phase using a full coefficient updating unit with decision feedback. This segmentation allows each phase to be optimized independently for its specific requirements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent dynamically switches between different coefficient updating units based on the equalization stage. The switching mechanism allows the system to adapt its structure: using the simplified unit during blind equalization when decision accuracy is not available, and transitioning to the full unit when decision-directed feedback becomes reliable.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If the iterative update step parameter or iterative update algorithm of the coefficient updating unit is modified to improve convergence error, then convergence error is improved, but convergence speed deteriorates

Engineering Contradiction:
Improveconvergence errorVSAvoidconvergence speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The patent divides the equalization process into two segments: blind equalization using a simplified updating unit, and decision-directed equalization using a full updating unit. This allows the system to achieve fast initial convergence through the simplified unit, then refine precision using the full unit's decision feedback capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary blind equalization using the simplified coefficient updating unit to establish an initial working state before transitioning to decision-directed equalization. This preliminary action prepares the system for the more precise but slower decision-directed phase.

Inventive Principle:
Principle #10Preliminary action

3Speed

If the coefficient updating unit is optimized for convergence speed or convergence error, then those specific performance metrics are improved, but blind equalization convergence capability cannot be improved

Engineering Contradiction:
Improveconvergence speedVSAvoidblind equalization convergence capability
Core Design Contradiction:
SpeedVSAdaptability or versatility

Solution Approach 1:

The patent implements a universal equalization system that can perform both blind equalization and decision-directed equalization by switching between two coefficient updating units. The first unit is optimized for blind equalization convergence, while the second unit provides full functionality for decision-directed operation, making the overall system versatile.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent dynamically adapts the equalization approach based on operational conditions. The switching mechanism between different coefficient updating units allows the system to optimize its blind equalization convergence capability when operating in blind mode, while maintaining full decision-directed capability when conditions permit.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2613488B1Adaptive equalization method and adaptive equalizer
Publication Date: 2016.01.20 HUAWEI TECH CO LTD
  • EP2613488B1 patent drawingFigure 1
  • EP2613488B1 patent drawingFigure 2
  • EP2613488B1 patent drawingFigure 3

AI summary

The present invention relates to the field of network communication, and specifically discloses an adaptive equalization method, including: obtaining a first filter signal according to a first filter coefficient; deciding the first filter signal based on an original constellation map to obtain a first decision signal, and deciding the first filter signal based on a level (n - 1) constellation map to obtain a level (n - 1) pseudo decision signal; comparing the level (n - 1) pseudo decision signal with the first filter signal to obtain a level (n - 1) error signal; if average energy of the level (n - 1) error signal is smaller than a level (n - 1) threshold, switching the level (n - 1) constellation map to a level n constellation map; obtaining an update magnitude of the filter coefficient according to the level (n - 1) error signal and the buffer signals, and obtaining a second filter coefficient according to the update magnitude; obtaining a second filter signal according to the second filter coefficient; and deciding the second filter signal based on the original constellation map to obtain a second decision signal. Embodiments of the present invention also disclose an adaptive equalizer.