PAM-4 Symbol Error Mapping from FEC Bit Error Vectors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

High-speed data link standards like 400 Gigabit Ethernet face challenges in symbol error analysis due to the lack of detailed symbol-level error information, which is unavailable because forward error correction operates at the bit level, and the use of unframed test signals that do not approximate real-world framed signals, making proper link tuning difficult.

Innovation Solution

An apparatus and method for symbol error analysis based on bit error vector evaluation, which demultiplexes 4-level PAM-4 lanes into NRZ lanes, performs forward error correction, and generates symbol error information by mapping bit errors to corresponding symbols, enabling detailed symbol error analysis and link tuning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If forward error correction is implemented at the bit level, then bit errors can be corrected, but symbol-level error information becomes unavailable

Engineering Contradiction:
Improvebit error correction capabilityVSAvoidsymbol-level error information
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent segments the bit error vector information into symbol-level error information by grouping bit errors that correspond to the same symbol. The symbol error information generator divides the bit error vector into multiple sections, where each section corresponds to errors in a specific symbol, thereby recovering symbol-level error details from bit-level correction data.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary component (symbol error information generator) that acts as a bridge between the bit-level forward error correction process and symbol-level analysis requirements. This intermediary processes the bit error vector to generate symbol error information, enabling both bit-level correction and symbol-level monitoring without direct interference between the two functions.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If unframed test signals are used for link testing, then testing can be performed, but the results do not accurately represent real-world framed signal performance

Engineering Contradiction:
Improvetesting capabilityVSAvoidlink quality assessment accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent changes the test signal parameters from unframed formats (PRBS, PRBSQ) to framed signal formats that match real-world transmission conditions. By modifying the signal structure to include framing overhead and data portions similar to actual Ethernet or storage protocols, the measurement accuracy improves while maintaining testing productivity.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If transmitter predistortion and receiver equalization are implemented, then signal impairments can be reduced, but detailed symbol error information is still unavailable for proper tuning

Engineering Contradiction:
Improvesignal quality improvementVSAvoidlink tuning capability
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent implements a feedback mechanism where symbol error information is generated from bit error vectors and fed back to the tuning system. This feedback loop provides detailed information about symbol-level errors, enabling operators to properly adjust predistortion and equalization parameters based on actual error patterns rather than attempting blind optimization.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3751778A1Evaluation of bit error vectors for symbol error analysis
Publication Date: 2020.12.16 VIAVI SOLUTIONS INC(US)
  • EP3751778A1 patent drawingFigure 1
  • EP3751778A1 patent drawingFigure 2
  • EP3751778A1 patent drawingFigure 3

AI summary

The disclosure relates to evaluating bit error vectors for symbol error analysis on real-world framed signals. Forward error correction (FEC) may generate a bit error vector to correct binary lanes such as non-return-to-zero (NRZ) lanes demultiplexed from a symbol-encoding lane such as a 4-level pulse amplitude modulation (PAM-4) lane. An apparatus may apply the bit error vector to the demultiplexed NRZ lanes to identify bit errors that occurred on the NRZ lanes. The apparatus may map the bit errors on the NRZ lanes to symbol errors on the PAM-4 lane. The apparatus may generate detailed symbol error information based on the identified symbol errors. The symbol error information may then be used for link tuning, thereby mitigating the effects of high frequency physical effects and other impairments on high-speed data links.