Decision Feedback Equalizer for High-Speed Wireline ISI Mitigation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Maintaining signal integrity in high-speed communication channels is challenging due to inter-symbol-interference (ISI), which worsens with increasing data rates and channel lengths, particularly in copper-based and optical fiber transmissions.

Innovation Solution

The implementation of a decision feedback equalizer (DFE) that estimates and subtracts inter-symbol-interference from received signals using a least mean squares (LMS) error algorithm, combined with parallel and pseudo-static dynamic architectures to reduce computational complexity and meet high data throughput requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data rate and channel length are increased, then transmission capacity is improved, but inter-symbol-interference (ISI) worsens

Engineering Contradiction:
Improvedata rateVSAvoidinter-symbol-interference
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent implements decision feedback equalization where the receiver uses feedback from previously detected symbols to estimate and subtract ISI from the current received signal. The DFE filter computes ISI based on detected symbols and feeds this back to the slicer, enabling the system to maintain signal integrity at high data rates by continuously compensating for interference.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent dynamically adjusts equalization parameters including DFE filter coefficients and slicer thresholds to optimize performance at different data rates. The system changes operational parameters based on channel conditions and transmission rate to mitigate ISI effects while maintaining high productivity.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If conventional equalization is used to compensate for ISI, then signal integrity is improved, but computational complexity increases

Engineering Contradiction:
Improvesignal integrityVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the equalization computation into separate functional blocks: a DFE filter for ISI estimation, adders for combining signals, and slicers for decision-making. This segmentation allows each block to perform a specific function with optimized complexity, reducing overall computational burden while maintaining signal integrity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent employs dynamic architectures including pipeline registers and clocking mechanisms that adapt the computational节奏 to match the data rate. The system dynamically manages computation timing and resource allocation to reduce complexity while preserving equalization effectiveness at high speeds.

Inventive Principle:
Principle #15Dynamics

3Productivity

If high data rates are implemented, then transmission capacity is improved, but timing requirements become more stringent

Engineering Contradiction:
Improvedata throughputVSAvoidtiming requirements
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent uses pipeline registers to pre-compute and store intermediate results before they are needed in subsequent stages. This preliminary action allows computations to be staged and executed in advance, reducing critical path delays and meeting stringent timing requirements at high data rates.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent employs clocked sequential logic with periodic operation cycles that synchronize data movement and computation. This periodic action ensures that all operations complete within each clock cycle, maintaining timing discipline while achieving high throughput through efficient clocked architecture.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentUS20240121139A1DFE implementation for wireline applications
Publication Date: 2024.04.11 TEXAS INSTRUMENTS INC
  • US20240121139A1 patent drawing
  • US20240121139A1 patent drawing
  • US20240121139A1 patent drawing

AI summary

Disclosed embodiments include a decision feedback equalizer (DFE) comprising an N-bit parallel input adapted to be coupled to a communication channel and configured to receive consecutive communication symbols, a first DFE path including a first path input configured to receive communication symbols, and a first adder having a first adder input coupled to the first path input. There is a first DFE filter having outputs responsive to the first DFE filter inputs, the outputs coupled to the second adder input. The DFE includes a first path having a first slicer and a first multiplexer, a first path multiplexer output, and a second DFE path including a second path input configured to receive a second communication symbol, a second adder, a second DFE filter, a second slicer, and a second multiplexer.