Equal-Length Vector Segmentation for Stable Gain Quantization

Solution Overview

Problem

Conventional methods for positional coding of audio/video signals face inefficiencies due to large codeword indices and varying energy content across input vector segments, leading to unstable gain quantization and inefficient coding.

Innovation Solution

A method for non-recursive segmentation of input vectors into equal-sized segments, followed by recursive determination of relative energy differences between segments, allowing for efficient coding by distributing bits based on energy and segment length.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If conventional clustering is used to reduce complexity, then device complexity is reduced, but manufacturing precision deteriorates because different parts of the input vector have very different sizes making positional coding inefficient

Engineering Contradiction:

Improveclustering complexityVSAvoidcoding precision

Core Design Contradiction:

Device complexityVSManufacturing precision

Solution Approach 1:

The input vector is divided into multiple segments of equal length, where each segment is processed separately through clustering and positional coding. This segmentation ensures that each segment maintains manageable size for efficient coding while collectively representing the entire input vector, thus resolving the contradiction between complexity reduction and coding precision.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different clustering operations are performed on different segments of the input vector, allowing each segment to be processed with appropriate local characteristics. This local processing approach maintains coding precision within each segment while keeping overall device complexity manageable through modular processing.

Inventive Principle:

Principle #3Local quality

2Adaptability or versatility

If the input vector is segmented into unequal parts, then adaptability improves for handling varying energy content, but manufacturing precision deteriorates because positional coding becomes inefficient

Engineering Contradiction:

Improveenergy distribution adaptabilityVSAvoidpositional coding precision

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The input vector is divided into multiple segments of equal length, where each segment is processed separately through clustering and positional coding. This segmentation ensures that each segment maintains manageable size for efficient coding while collectively representing the entire input vector, thus resolving the contradiction between complexity reduction and coding precision.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The energy information of each segment is represented through parameters such as energy ratios and differential energy values. By transforming the energy distribution characteristics into these parameters, the system can adapt to varying energy content while maintaining equal segment lengths for efficient positional coding.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If conventional positional coding is applied to long input vectors, then measurement precision is maintained, but productivity deteriorates due to rapidly increasing number of combinations by increasing dimensions

Engineering Contradiction:

Improvepositional coding accuracyVSAvoidcoding efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The input vector is divided into multiple segments of equal length, where each segment is processed separately through clustering and positional coding. This segmentation ensures that each segment maintains manageable size for efficient coding while collectively representing the entire input vector, thus resolving the contradiction between complexity reduction and coding precision.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The problem of high-dimensional positional coding is transformed by introducing a new dimension of segmentation. Instead of directly coding positions in a high-dimensional space, the system divides the space into multiple lower-dimensional segments, codes each segment separately, and then combines the results, effectively reducing the computational complexity while maintaining coding accuracy.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS11621725B2Methods and devices for vector segmentation for coding

Publication Date: 2023.04.04 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

A method for partitioning of input vectors for coding is presented. The method comprises obtaining of an input vector. The input vector is segmented, in a non-recursive manner, into an integer number, NSEG, of input vector segments. A representation of a respective relative energy difference between parts of the input vector on each side of each boundary between the input vector segments is determined, in a recursive manner. The input vector segments and the representations of the relative energy differences are provided for individual coding. Partitioning units and computer programs for partitioning of input vectors for coding, as well as positional encoders, are presented.