Input Vector Segmentation for Stable Gain Quantization Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional positional coding schemes face inefficiencies due to large codeword indices and varying sizes of input vector segments, leading to unstable gain quantization and inefficient coding, especially for long input vectors.
Innovation Solution
A method and device for non-recursive segmentation of input vectors into equal-sized segments, with recursive determination of relative energy differences between segments, enabling efficient positional coding by distributing bits based on energy and segment lengths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional clustering is used to reduce complexity, then implementation complexity is reduced, but coding efficiency deteriorates due to highly variable segment sizes
Solution Approach 1:
The input vector is divided into multiple segments, where each segment is further divided into sub-segments. This hierarchical segmentation allows the system to manage complexity by breaking down large vectors into smaller, more manageable parts while maintaining coding efficiency through controlled segment size variation.
Solution Approach 2:
Different segments are allowed to have different sizes based on local energy characteristics. High-energy segments can be larger to capture more information, while low-energy segments are smaller. This local adaptation optimizes coding efficiency without requiring complex global restructuring.
2Productivity
If input vectors are segmented into equal sizes, then coding efficiency is improved through stable gain quantization, but flexibility in handling varying energy content deteriorates
Solution Approach 1:
The segment size is made dynamic rather than fixed. Each segment's size is adjusted based on its energy content, allowing the segmentation scheme to adapt to varying signal characteristics while maintaining the benefits of structured segmentation for coding efficiency.
Solution Approach 2:
The segment size parameter is changed based on energy measurements. By calculating energy in potential segments and comparing against thresholds, the system dynamically adjusts segment boundaries to optimize both coding efficiency and adaptability to different signal conditions.
3Device complexity
If non-recursive segmentation is used, then processing complexity is reduced, but ability to handle varying segment energies deteriorates
Solution Approach 1:
Energy calculations and segment boundary decisions are made in advance during the segmentation phase, before actual coding begins. This preliminary action allows the system to prepare optimal segmentations based on energy characteristics without adding complex processing during the coding stage itself.
Solution Approach 2:
Energy thresholds act as intermediaries between the raw signal and the segmentation decision. By comparing segment energies against predefined thresholds, the system can make informed segmentation decisions without requiring complex recursive analysis, thus reducing processing complexity while maintaining energy awareness.
Data Source
Figure 1A~4
Figure 2
Figure 5~6
AI summary
A method for partitioning of input vectors for coding is presented. The method comprises obtaining (210) of an input vector. The input vector is segmented (220), 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 (230), in a recursive manner. The input vector segments and the representations of the relative energy differences are provided (250) for individual coding. Partitioning units and computer programs for partitioning of input vectors for coding, as well as positional encoders, are presented.