Hierarchical Data Decoding via Mixed-Class Error Protection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing error correction encoding methods, such as turbo codes and UEP codes, face challenges in optimizing digital data transmission resources, leading to increased transmission and reception delays, higher network capacity requirements, and resource inefficiencies due to separate processing of data classes.

Innovation Solution

A method that encodes digital data into classes with varying levels of protection by mixing and re-encoding data from preceding steps, using techniques like interleaving and recursive systematic convolutional codes, to achieve hierarchical error protection, reducing the need for separate processing and modulation schemes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If UEP codes process each class separately with separate encoding and modulation, then hierarchical error protection is achieved, but transmission delays increase and network resources are wasted due to need for headers and resynchronization

Engineering Contradiction:
Improveerror protectionVSAvoidtransmission delay
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent merges multiple data classes into a single encoded stream by mixing data from different classes before encoding. The receiver mixes received data from all classes and decodes them together in a unified decoding process, eliminating the need for separate processing of each class while maintaining hierarchical error protection through the mixing mechanism.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces a mixing operation as an intermediary step between data collection and encoding. This mixing operation combines data from multiple classes into a unified stream that can be processed together, serving as a mediator that enables joint encoding while preserving the hierarchical structure through the mixing pattern.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If UEP codes process each class separately with separate encoding and modulation, then hierarchical error protection is achieved, but network capacity and resources increase due to need for headers and resynchronization processing

Engineering Contradiction:
Improveerror protectionVSAvoidnetwork capacity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges multiple data classes into a single encoded stream by mixing data from different classes before encoding. The receiver mixes received data from all classes and decodes them together in a unified decoding process, eliminating the need for separate processing of each class while maintaining hierarchical error protection through the mixing mechanism.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent extracts the hierarchical error protection function from the encoding structure and embeds it in the mixing operation. By taking out the class separation from the encoding process and placing it in the mixing domain, the system achieves hierarchical protection without requiring separate encoding structures, headers, or resynchronization mechanisms.

Inventive Principle:
Principle #2Taking out (Extraction)

3Productivity

If all data classes are encoded and modulated together without separate processing, then transmission delays and network capacity requirements are reduced, but hierarchical error protection may be compromised

Engineering Contradiction:
Improvetransmission rateVSAvoiderror protection
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent applies local quality by making different data classes contribute differently to the mixed encoded stream through the mixing operation. The mixing pattern assigns different weights or positions to data from different classes, providing stronger implicit protection to higher priority classes while maintaining unified encoding and modulation for all classes.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent introduces asymmetry in the mixing operation where data from different classes are combined in non-uniform patterns. This asymmetric mixing provides differential error protection without requiring separate encoding schemes, as the mixing structure itself creates unequal protection levels for different classes within a unified encoding framework.

Inventive Principle:
Principle #4Asymmetry

Data Source

PatentEP2686964B1Decoding method, decoding device and corresponding computer program product
Publication Date: 2022.06.01 AIRBUS DEFENCE & SPACE SAS
  • EP2686964B1 patent drawingFigure 1~2
  • EP2686964B1 patent drawingFigure 3~4
  • EP2686964B1 patent drawingFigure 5~6

AI summary

The present invention relates to an error correction encoding method (100) for encoding so called source digital data (30) having the form of a frame (102), wherein said data can be classified into a plurality of classes (102i). The encoding method according to the invention comprises the following steps: - a first encoding step (1101) for encoding data to be encoded formed by the data of a first class (1021), to obtain encoded data; and - implementing the following step successively for at least one other class (102i>1): - mixing (108i>1) data of said other class (102i>1) and the data encoded or to be encoded from a preceding encoding step, and - encoding (110i>1) data to be encoded formed by said mixed data to obtain encoded data. The invention also relates to a method for decoding data encoded with the encoding method according to the invention, as well as associated encoding device and decoding device.