Signal Receiver Symbol Distribution Error Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current communication systems face inefficiencies due to shaping losses in AWGN channels, particularly for short block lengths, where probabilistic shaping schemes like CCDM incur rate losses and iterative decoding performs poorly, leading to suboptimal performance.

Innovation Solution

A signal receiver is configured to decode signals by comparing their symbol frequency with a predetermined distribution, requesting retransmissions or selecting candidate representations based on matching frequencies, and employing maximum likelihood detection to improve decoding accuracy, thereby reducing overhead and rate loss.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If probabilistic amplitude shaping (PAS) with constant composition distribution matching (CCDM) is used to approach channel capacity, then shaping gain is achieved, but rate loss occurs for short block lengths that could outweigh the shaping gain

Engineering Contradiction:
Improvechannel capacity approachVSAvoidrate loss
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies feedback by using the known symbol distribution as a reference to verify decoding correctness. The receiver compares the actual symbol frequencies in the decoded sequence against the expected distribution from CCDM, and uses this feedback to detect errors and request retransmissions when mismatches occur, thereby improving reliability without sacrificing rate.

Inventive Principle:
Principle #23Feedback

2Ease of operation

If iterative decoding is used for short block lengths, then decoding is performed, but performance is poor

Engineering Contradiction:
Improvedecoding operationVSAvoiddecoding performance
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent introduces an intermediary error detection mechanism based on symbol distribution verification. This intermediary check sits between the iterative decoding process and the final decision, using the known CCDM distribution as a mediator to verify whether the decoded sequence is valid, thereby improving reliability without adding complex iterative procedures.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If outer error detecting codes such as CRC are added to select valid codewords, then error detection capability is improved, but overhead and rate loss increase significantly for short blocks

Engineering Contradiction:
Improveerror detection capabilityVSAvoidoverhead and rate loss
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent makes the system self-service by using the inherent symbol distribution property of CCDM-encoded signals as its own error detection mechanism. Instead of adding external CRC codes, the system uses the known distribution characteristics embedded in the encoding process itself to verify correctness, thereby achieving error detection without additional overhead or rate loss.

Inventive Principle:
Principle #25Self-service

4Ease of manufacture

If uniform QAM symbols are used, then implementation is simple, but shaping loss of up to 1.53 dB occurs for high order constellations

Engineering Contradiction:
Improveimplementation simplicityVSAvoidshaping loss
Core Design Contradiction:
Ease of manufactureVSLoss of energy

Solution Approach 1:

The patent applies parameter changes by modifying the symbol distribution parameters from uniform to the specific discrete distribution required by CCDM. This changes the statistical parameters of the QAM symbols to match the optimal distribution for the channel, thereby reducing shaping loss while maintaining practical implementability through the use of standard QAM constellations with adjusted symbol selection probabilities.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3504814B1Error detection using symbol distribution in a system with distribution matching and probabilistic amplitude shaping
Publication Date: 2021.05.26 HUAWEI TECH DUESSELDORF
  • EP3504814B1 patent drawingFigure 1~2
  • EP3504814B1 patent drawingFigure 3~4
  • EP3504814B1 patent drawingFigure 5~7

AI summary

A signal receiver for interpreting a received signal, the receiver being configured to perform the steps of: decoding the received signal so as to form a sequence of symbols hypothesised to represent the content of the received signal; comparing the frequency of occurrence of symbols within the sequence with a predetermined symbol distribution; and if the relative frequency of occurrence of symbols within the sequence of symbols does not match the predetermined distribution, treating the sequence of symbols as being an incorrect representation of the content of the received signal.