Decoder State Resynchronization After Audio Packet Loss Concealment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing packet loss concealment methods are inadequate for sub-band predictive coders like the ITU-T G.722 wideband speech coder, particularly due to structural issues and challenges posed by Adaptive Differential Pulse Code Modulation (ADPCM) predictive coders, which are not addressed by prior-art techniques.

Innovation Solution

A method and system for updating the decoder state in a predictive coding system by re-encoding the synthesized audio signal associated with a lost frame using a simplified encoder, and updating the decoder state based on the generated encoder state to address sub-band-specific architectural issues and backward-adaptive ADPCM challenges.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If prior-art packet loss concealment methods are applied to sub-band predictive coders, then packet loss effects can be concealed, but sub-band-specific structural issues and backward-adaptive ADPCM challenges are not addressed, resulting in poor audio quality

Engineering Contradiction:
Improvepacket loss concealment effectivenessVSAvoidadaptability to sub-band predictive coder architecture
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent applies local quality by treating each sub-band independently with sub-band-specific packet loss concealment processing. The decoder state update mechanism is specifically adapted for each sub-band's ADPCM structure, applying localized corrections to predictor coefficients and quantizer step sizes rather than uniform processing across all bands.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes parameters by dynamically updating decoder internal states (predictor coefficients, quantizer step sizes) to match encoder states after packet loss. The method adjusts these parameters based on re-encoded synthesized audio, ensuring the decoder operates with correct state values that adapt to the actual signal characteristics.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If a full encoder is used to re-encode synthesized audio for decoder state update, then accurate state alignment is achieved, but computational complexity and processing time increase significantly

Engineering Contradiction:
Improvedecoder state alignment accuracyVSAvoidencoder processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the necessary encoder components needed for state update, rather than implementing a complete encoder. Specifically, only the sub-band analysis filter bank and ADPCM encoding logic are used to generate state information, while other encoder functions are omitted. This extraction reduces complexity while maintaining the essential state alignment capability.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary action by pre-calculating and storing encoder state information during normal operation. When packet loss occurs, the system uses these pre-computed states rather than performing full re-encoding, significantly reducing the computational burden during loss concealment while maintaining accuracy.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If decoder state is not updated after packet loss concealment, then processing is simpler, but internal state differences between encoding and decoding accumulate, causing audio quality degradation

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidaudio quality
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements feedback by using the synthesized concealed audio as input to the simplified encoder, then using the resulting encoder state to update the decoder state. This closed-loop feedback mechanism ensures the decoder state continuously aligns with what the encoder would produce, maintaining long-term synchronization and audio quality despite periodic packet losses.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent applies self-service by having the decoder update its own state using information generated from its own synthesized output. The decoder synthesizes concealed audio, feeds it through a simplified encoder model, and uses the resulting state to correct its internal registers, making the system self-correcting without external intervention.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2054877B1Updating of decoder states after packet loss concealment
Publication Date: 2011.10.26 BROADCOM INC
  • EP2054877B1 patent drawingFigure 1
  • EP2054877B1 patent drawingFigure 2
  • EP2054877B1 patent drawingFigure 3

AI summary

A technique is described herein for updating a state of a decoder in a predictive coding system after synthesizing an audio output signal corresponding to a lost frame in a series of frames representing an encoded audio signal. In accordance with the technique, an audio signal associated with the synthesized output audio signal is re-encoded in an encoder to generate an encoder state, wherein the encoder is simplified with respect to an encoder used to generate the encoded audio signal. The state of the decoder is then updated based on the generated encoder state.