MDCT Error Concealment via Spectral Bin Sign Assignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio coding and decoding techniques, such as those using Modified Discrete Cosine Transforms (MDCT), face challenges in error concealment due to the trade-off between quality and complexity, leading to audible distortions from packet errors or losses.
Innovation Solution
A method and system for error concealment in MDCT-based audio decoders that generate estimated MDCT coefficients by assigning signs to tonal-like and noise-like spectral bins based on preceding packets, creating a concealment packet to replace erroneous packets, while maintaining low complexity and minimizing distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If estimating concealment methods are used to replace erroneous frames with estimations, then error concealment quality is improved, but computational complexity increases
Solution Approach 1:
The patent segments the spectral bins into different types (tonal-like and noise-like) and applies different sign assignment strategies to each segment. This segmentation allows the system to achieve good error concealment quality for tonal components while using simpler methods for noise components, thereby balancing quality and complexity
Solution Approach 2:
The patent changes the parameter of sign assignment from random (prior art) to deterministic based on spectral bin characteristics. By changing how signs are assigned (from random to structured), the patent improves error concealment quality while maintaining computational efficiency through a more intelligent parameter selection strategy
2Manufacturing precision
If complex estimation methods are applied to improve error concealment, then audio quality is improved, but processing complexity increases
Solution Approach 1:
The patent applies different processing strategies to different local regions of the spectrum. Tonal spectral bins receive deterministic sign assignment based on neighboring bins, while noise-like bins receive random sign assignment. This local differentiation improves overall audio quality without requiring complex processing across the entire spectrum
Solution Approach 2:
The patent performs preliminary classification of spectral bins into tonal-like and noise-like categories before error concealment. This preliminary action allows the system to prepare appropriate sign assignment strategies in advance, improving error concealment quality while avoiding complex real-time processing during the concealment operation itself
Data Source
AI summary
An error-concealing audio decoding method comprises: receiving a packet comprising a set of MDCT coefficients encoding a frame of time-domain samples of an audio signal; identifying the received packet as erroneous; generating estimated MDCT coefficients to replace the set of MDCT coefficients of the erroneous packet, based on corresponding MDCT coefficients associated with a received packet directly preceding the erroneous packet; assigning signs of a first subset of MDCT coefficients of the estimated MDCT coefficients, wherein the first subset comprises such MDCT coefficients that are associated with tonal-like spectral bins, to coincide with signs of corresponding MDCT coefficients of said preceding packet; randomly assigning signs of a second subset of MDCT coefficients of the estimated MDCT coefficients, wherein the second subset comprises MDCT coefficients associated with noise-like spectral bins; replacing the erroneous packet by a concealment packet containing the estimated MDCT coefficients and the signs assigned.