128b/132b Block Header Encoding for Header Error Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing line coding schemes, such as 8b/10b and 64b/66b, face challenges in detecting and correcting errors in block headers during data transmission, particularly in ensuring reliable recovery of data payloads over physical media.

Innovation Solution

The implementation of a 128b/132b line coding scheme with a 4-bit block header that achieves a hamming distance of at least four between data and control block headers, enabling single-bit error correction and double-bit error detection, ensuring reliable data transmission and recovery.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing line coding schemes (8b/10b, 64b/66b) are used, then data transmission is achieved, but error detection and correction capability in block headers is insufficient

Engineering Contradiction:
Improveerror detection and correction capabilityVSAvoidcoding scheme complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent changes the parameters of the block header by increasing its length to 4 bits and specifically designing the bit patterns for data blocks (0xxx where x≠0) and control blocks (1xxx where x≠0). This parameter change enables the achievement of a minimum Hamming distance of 4 between data and control block headers, providing robust error detection and correction capability while maintaining a manageable coding scheme structure.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If block header error correction is enhanced, then data recovery reliability improves, but the coding scheme becomes more complex

Engineering Contradiction:
Improvedata recovery reliabilityVSAvoidline coding scheme complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-defining specific bit patterns for data block headers (0xxx where x≠0) and control block headers (1xxx where x≠0) during the encoding phase. This preliminary design of the header structure with built-in error detection properties allows the receiver to automatically detect and correct errors without requiring complex real-time processing, thus improving data recovery reliability while controlling scheme complexity.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If a 4-bit block header with Hamming distance of at least four is implemented, then single-bit error correction and double-bit error detection are enabled, but the encoding overhead increases

Engineering Contradiction:
Improveerror correction capabilityVSAvoidheader overhead
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent applies local quality by concentrating the error protection mechanism specifically in the block header portion of the data structure. The 4-bit header is designed with specific bit patterns (0xxx for data blocks, 1xxx for control blocks) that provide Hamming distance of at least 4, enabling error detection and correction. This localized approach provides robust error protection where it is most needed for data recovery while minimizing the overall overhead compared to applying similar protection to the entire data payload.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS10997016B2Method of encoding data
Publication Date: 2021.05.04 INTEL CORP
  • US10997016B2 patent drawing
  • US10997016B2 patent drawing
  • US10997016B2 patent drawing

AI summary

Techniques for encoding data are described herein. The method includes receiving a block payload at a physical layer to be transmitted via a data bus. The method includes establishing a block header comprising an arrangement of bits, the block header defining two block header types, wherein a hamming distance between block header types is at least four.