Bandwidth Extension Signal Generation via Anti-Sparseness Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio encoding methods struggle to efficiently generate bandwidth extended signals for high-frequency bands, leading to metallic noise issues due to insufficient encoding and decoding techniques.
Innovation Solution
The method involves performing anti-sparseness processing on low-frequency spectra and high-frequency extension encoding in the frequency domain to reduce metallic noise and enhance the quality of high-frequency extended signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If spectral band replication (SBR) is used to encode high-frequency bands using parameters from low-frequency bands, then coding efficiency is improved and bit allocation is optimized, but metallic noise is generated in the bandwidth extended signal
Solution Approach 1:
The patent applies parameter changes by modifying the spectral characteristics of the low-frequency band through anti-sparseness processing. This involves changing the spectral envelope parameters and adding spectral components to prevent the generation of metallic noise in the high-frequency extended band, while maintaining the efficiency benefits of SBR encoding
Solution Approach 2:
The patent introduces an intermediary processing step (anti-sparseness processing) between the low-frequency encoding and high-frequency extension. This intermediary process modifies the low-frequency spectrum to prevent noise generation during bandwidth extension, acting as a mediator that resolves the conflict between encoding efficiency and noise reduction
2Quantity of substance
If a relatively small number of bits is allocated to high-frequency band encoding, then bit rate is reduced and transmission efficiency is improved, but the quality of the bandwidth extended signal deteriorates with metallic noise
Solution Approach 1:
The patent applies preliminary action by performing anti-sparseness processing on the low-frequency spectrum before bandwidth extension. This preprocessing step prepares the spectral data in advance to prevent metallic noise generation during high-frequency reconstruction, ensuring signal quality without requiring additional bits for high-frequency encoding
3Device complexity
If bandwidth extension is performed using correlation between low and upper bands, then coding complexity is reduced, but spectrum holes are generated causing metallic noise in the extended high-frequency signal
Solution Approach 1:
The patent modifies spectral parameters through anti-sparseness processing to fill spectrum holes that would otherwise be created during bandwidth extension. By changing the spectral density and adding components to the low-frequency representation, the method prevents metallic noise while maintaining the simplicity of correlation-based extension
Solution Approach 2:
The patent converts the potential harm of spectrum holes into a benefit by using anti-sparseness processing to intentionally add spectral components. This transforms what would be empty spectrum regions (causing noise) into filled regions that prevent noise generation, while still maintaining the low-complexity bandwidth extension approach
Data Source
AI summary
An apparatus for generating a bandwidth extended signal includes an anti-sparseness processing unit to perform anti-sparseness processing on a low-frequency spectrum; and a frequency domain high-frequency extension decoding unit to perform high-frequency extension encoding in the frequency domain on the low-frequency spectrum on which the anti-sparseness processing is performed.