Audio Codec Permutation Coding With Sorted Amplitude Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio coding methods, such as those in the G.718 standard, face increased storage complexity due to the need to store vectors representing element removal values and hierarchical combination coding parameters, especially for lattice vector quantizers with many bits and codewords.
Innovation Solution
A coding method based on a Gosset lattice that simplifies storage by obtaining and sorting amplitude and length vectors in real time, allowing for permutation coding without storing the vectors representing element removal values and hierarchical combination coding parameters, thereby reducing storage complexity and improving performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If hierarchical combination coding method is used, then coding performance is improved, but storage complexity increases
Solution Approach 1:
The patent extracts only the essential coding parameters (amplitude vector and length vector) from the hierarchical combination coding method, discarding the complex pre-stored vectors representing element removal values and hierarchical combination coding parameters. This extraction maintains coding performance while significantly reducing storage complexity.
Solution Approach 2:
The patent changes the parameter representation from storing complex vectors (element removal values and hierarchical combination coding parameters) to storing simplified parameters (amplitude vector and length vector). This parameter transformation achieves the same coding functionality with reduced storage requirements.
2Manufacturing precision
If lattice vector quantizer with many bits and codewords is used, then coding quality is improved, but storage complexity increases significantly
Solution Approach 1:
The patent replaces expensive, long-term storage of large codebooks with cheaper, on-the-fly computation of amplitude and length vectors. The simplified parameters can be regenerated as needed without requiring extensive permanent storage, analogous to using disposable resources instead of permanent infrastructure.
Solution Approach 2:
The patent performs preliminary organization of the codebook structure into amplitude and length vectors that can be efficiently accessed and processed. This preliminary organization enables faster coding operations and reduces the need for storing additional auxiliary data structures.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A coding method, a decoding method, a coding-decoding (codec) method, a codec system and relevant apparatuses are disclosed. The coding method includes: obtaining an amplitude vector and a length vector corresponding to a vector to be coded; sorting elements of the amplitude vector and elements of the length vector; and obtaining a position index value according to the sorted amplitude vector and the sorted length vector. A decoding method, a codec system, and relevant apparatuses are also provided.