Correlated Noise Canceller for Ethernet Receivers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

High-speed Ethernet systems, such as 10 GBASE-T, face challenges in reducing correlated noise like radio frequency interference (RFI) without increasing package pincount by employing a fifth channel for common-mode noise detection.

Innovation Solution

A correlated noise canceller is integrated into the slicer circuit of the receiver, utilizing a feedback path with a finite impulse response (FIR) filter to predict and filter out RFI based on slicer error signals, enhancing signal fidelity without the need for an additional channel.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a fifth channel is employed for common-mode noise detection, then RFI filtering capability is improved, but package pincount increases

Engineering Contradiction:
ImproveRFI filtering capabilityVSAvoidpackage pincount
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent employs feedback by using the slicer error signal from the existing data channels to drive the correlated noise canceller. The canceller processes historical error signals to predict and cancel future noise, creating a closed-loop system that eliminates the need for a separate fifth channel for noise detection.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces an intermediary approach by using the slicer error signal as a mediator between the received signal and the noise cancellation process. Instead of directly detecting common-mode noise through a fifth channel, the system uses the error signal from the slicer as an intermediate representation to drive the correlated noise canceller, achieving noise filtering without additional physical channels.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If correlated noise canceller is added to slicer circuit, then signal fidelity is improved, but device complexity increases

Engineering Contradiction:
Improvesignal fidelityVSAvoidcircuit complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges the correlated noise canceller with the existing slicer circuit by integrating the noise cancellation functionality into the slicer's feedback path. The canceller processes the slicer error signal and feeds back corrected values, combining noise cancellation operations with the existing slicing and decision-making circuitry rather than adding completely separate processing paths.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The correlated noise canceller performs preliminary action by processing historical error signals to predict future noise patterns before they affect the current decision. The system uses past error signals to pre-compute cancellation values that will be applied to upcoming received signals, proactively preventing noise-induced errors rather than reacting to them after occurrence.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS8861663B1Correlated noise canceller for high-speed ethernet receivers
Publication Date: 2014.10.14 MARVELL ASIA PTE LTD
  • US8861663B1 patent drawing
  • US8861663B1 patent drawing
  • US8861663B1 patent drawing

AI summary

A method of operation in an ethernet receiver circuit is disclosed. The method comprises sampling an input signal to generate a sampled signal having a sampled noise component and a sampled data component. The sampled signal is sliced, and a slicer error determined based on the slicing of the sampled signal. A subsequently sampled noise component is filtered based on the slicer error.