Wideband Signal Encoding Reducing Processing Samples
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for encoding wideband speech/sound signals require excessive processing and increase information requirements, leading to severe sound quality degradation due to quantization distortion in the low band spectral data.
Innovation Solution
An encoding apparatus and method that utilize orthogonal transforms and integration of encoded information to reduce processing samples and maintain high sound quality, even with severe quantization distortion, by generating high band spectral data based on low band data through similar-part search and amplitude ratio adjustment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spectral data of the high band region is divided into a plurality of groups and the spectrum of the low band region that is the most similar to the spectrum of each group is regarded as side information, then high band signal generation is achieved, but the amount of processing calculations becomes enormous
Solution Approach 1:
The patent divides the high band signal into multiple frequency regions and processes each region separately using different strategies (copying from low band or generating from residual signal), rather than uniformly processing all high band data. This segmentation reduces overall processing complexity while maintaining spectral accuracy.
Solution Approach 2:
The patent applies different processing methods to different parts of the high band signal based on local characteristics. Specifically, some frequency regions are generated by copying low band spectra while others use residual signal generation, optimizing processing efficiency for each local region rather than applying a single method globally.
2Measurement precision
If processing is carried out on a per band basis to generate high band signal, then high band signal generation is achieved, but the amount of information required to encode side information increases
Solution Approach 1:
The patent discards detailed high band spectral information and recovers it by copying from low band regions or generating from compact residual signals. This approach significantly reduces the amount of side information that needs to be encoded and transmitted while still recovering the high band signal quality.
Solution Approach 2:
The patent copies spectral data from the low band region to generate the high band signal in certain frequency regions. This copying mechanism eliminates the need to encode and transmit separate high band spectral information, dramatically reducing side information requirements while maintaining signal quality.
3Measurement precision
If the degree of similarity between high band signal and low band signal is decided per subband, then high band signal generation is achieved, but not only the amount of calculation, but also the amount of information required to encode side information increases
Solution Approach 1:
The patent merges the processing of high band signal generation with the low band decoding process. By integrating the high band generation into the existing low band decoding framework, the patent avoids separate processing steps and reduces overall processing complexity while maintaining spectral accuracy.
Solution Approach 2:
The patent makes the low band decoded signal serve multiple functions: it is both the decoded output and the source for generating high band signals through copying and residual signal processing. This multi-functionality eliminates the need for separate high band processing systems, reducing computational complexity.
4Measurement precision
If spectral data of the low band region is used to generate high band signal, then high band signal generation is achieved, but severe sound quality degradation is anticipated when spectral data of the low band region is distorted by quantization
Solution Approach 1:
The patent dynamically selects between different high band generation methods (copying from low band or generating from residual signal) based on the quality and characteristics of the low band spectral data. This dynamic adaptation ensures high band signal quality even when low band data is distorted by quantization.
Solution Approach 2:
The patent introduces an intermediary processing stage that evaluates the quality of low band spectral data and mediates between the low band decoded signal and the high band signal generation. This intermediary layer prevents direct propagation of quantization distortion from low band to high band regions.
Data Source
AI summary
An encoding device and an encoding method are provided for encoding by reducing the number of samples to be processed when encoding higher-band spectrum data according to lower-band spectrum data in a wide-band signal. The device and the method can obtain a high-quality decoded signal even if a large quantization distortion is caused in the lower-band spectrum data. When encoding higher-band spectrum data in a signal to be encoded, according to lower-band spectrum data in the signal, only for a part (a head portion) of the higher-band spectrum data, the lower-band spectrum data after being quantized is subjected to approximate partial search and higher-band spectrum data is generated according to the search result.