Selective Slicing Equalizer for Stable Blind Coefficient Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In communication systems, blind equalizers face challenges with error propagation and noise enhancement when operating without a training sequence, leading to divergent coefficient values and mis-decision errors.
Innovation Solution
A selective slicing equalizer comprising a feed-forward filter, feedback filter, combiner, decision device, and control circuit that generates a slice control signal to select between combined and mapped signals, using multiple equalization algorithms to adjust coefficients and prevent error propagation, and employing region mapping to determine the slice control signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If blind equalization is used without training sequence, then the equalizer can operate without prior channel knowledge, but error propagation and noise enhancement occur leading to coefficient divergence
Solution Approach 1:
The patent implements a feedback mechanism where the equalized output is fed back through a feedback filter to cancel post-cursor intersymbol interference. The feedback filter coefficients are continuously updated based on the decision-directed error signal, creating a closed-loop system that stabilizes coefficient convergence during blind equalization operation.
Solution Approach 2:
The patent applies different processing to different parts of the signal: the feed-forward filter handles pre-cursor ISI with one set of adaptive coefficients, while the feedback filter handles post-cursor ISI with separate adaptive coefficients. This local differentiation allows each filter to optimize for its specific function, improving overall convergence reliability.
2Measurement precision
If slicing is applied to reduce error propagation, then decision accuracy improves, but noise enhancement occurs in blind mode
Solution Approach 1:
The patent dynamically adjusts the slicing operation based on the equalizer's operational state. During blind equalization mode, slicing is selectively applied or avoided depending on convergence status, while in decision-directed mode, slicing is consistently applied. This dynamic control allows the system to optimize decision accuracy while managing noise enhancement through state-dependent processing.
Solution Approach 2:
The patent segments the equalization process into distinct phases: blind equalization phase where slicing is controlled to prevent noise enhancement, and decision-directed phase where slicing is fully applied for maximum decision accuracy. This temporal segmentation allows the system to benefit from slicing when safe and avoid it when harmful.
Data Source
AI summary
An equalizer. The equalizer, either operated in a blind mode or a decision directed mode, comprises a feed-forward filter, a feedback filter, a decision device, a control circuit, and a multiplexer. The feed-forward filter receives an input signal. The feedback filter filters an equalized signal. The combiner combines the feed-forward filtered signal and the feedback filtered signal. The decision device maps the combined signal to one symbol of a symbol set. The control circuit receives the combined output and generates a slice control signal. The multiplexer selects the combined signal or the mapped signal as the equalized according to the slice control signal when operated in the blind mode.