Entropy Coding Context Transfer for Audio Resolution Switching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional audio coding concepts face inefficiencies when switching between different time/frequency resolutions, leading to sub-optimal coding efficiency due to context resetting, which is inadequate for frequent switching scenarios required for lower data rates and adaptive signal processing.

Innovation Solution

A context mapping mechanism is employed in context-based entropy coding to derive new contexts from old resolutions using interpolation, extrapolation, sub-sampling, and up/down-sampling, allowing for more frequent switching between time/frequency resolutions without compromising coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If context is reset when switching between different time/frequency resolutions, then synchronization between encoder and decoder is maintained, but coding efficiency deteriorates significantly

Engineering Contradiction:
ImprovesynchronizationVSAvoidcoding efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent copies the context state from the previous resolution to the current resolution, allowing the decoder to reconstruct the same context without resetting. This copying mechanism maintains synchronization while preserving coding efficiency by avoiding the loss of contextual information that would occur with a full reset.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent changes the context parameters dynamically based on the current resolution being used. Instead of resetting context to default values, the context parameters are adapted to match the current resolution while maintaining continuity with previous resolutions, thereby preserving coding efficiency across resolution switches.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If block length switching is performed frequently to adapt to signal characteristics, then adaptability improves, but coding efficiency deteriorates due to repeated context resets

Engineering Contradiction:
Improvesignal adaptationVSAvoidcoding efficiency
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent implements a dynamic context management system where the context state is continuously updated and carried forward across resolution switches. This dynamic approach allows frequent block length switching for signal adaptation while maintaining coding efficiency, as the context adapts smoothly rather than resetting repeatedly.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent ensures continuity of the context state across resolution switches, allowing the useful action of context-based coding to continue without interruption. By maintaining context continuity, the system can perform frequent block length switching for signal adaptation without the penalty of repeated context resets.

Inventive Principle:
Principle #20Continuity of useful action

3Productivity

If context is not reset during resolution switching, then coding efficiency is maintained, but synchronization between encoder and decoder may be lost

Engineering Contradiction:
Improvecoding efficiencyVSAvoidsynchronization
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the context state is explicitly managed and communicated between encoder and decoder. The decoder receives information about the current resolution and reconstructs the context state accordingly, ensuring synchronization is maintained while coding efficiency is preserved through continuous context usage.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3300076B1Audio encoder and audio decoder
Publication Date: 2019.04.24 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
  • EP3300076B1 patent drawingFigure 1
  • EP3300076B1 patent drawingFigure 2
  • EP3300076B1 patent drawingFigure 3

AI summary

An audio encoder (100) for encoding segments of coefficients, the segments of coefficients representing different time or frequency resolutions of a sampled audio signal, the audio encoder (100) comprising a processor (110) for deriving a coding context for a currently encoded coefficient of a current segment based on a previously encoded coefficient of a previous segment, the previously encoded coefficient representing a different time or frequency resolution than the currently encoded coefficient. The audio encoder (100) further comprises an entropy encoder (120) for entropy encoding the current coefficient based on the coding context to obtain an encoded audio stream.