Vector Quantization Coding With Normalization Error Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional coding methods fail to ensure a small error and improved signal-to-noise ratio (SNR) in decoding digital signals due to the reliance on normalization values obtained from input signals alone, leading to potential errors in the output signal.

Innovation Solution

A coding method that normalizes input signals, generates a correction coefficient to minimize the distance between the input and inverse-normalized signal sequences, and quantizes this coefficient along with the normalization value to produce a code including signal and normalization information indices, thereby correcting the normalization value for improved decoding accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If normalization value is obtained from input signal alone, then coding process is simple, but decoding accuracy deteriorates with larger errors

Engineering Contradiction:

Improvecoding process complexityVSAvoiddecoding accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces a feedback mechanism where the normalization value is corrected based on the quantized signal sequence. The correction value is calculated by comparing the quantized signal with the inverse-quantized signal, and this correction is fed back to improve the normalization value for subsequent coding, thereby reducing decoding errors while maintaining coding simplicity

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent replaces the traditional mechanical calculation of normalization value (based solely on input signal statistics) with a corrected normalization value that incorporates feedback from the quantization process. This substitution transforms the normalization value from a purely statistical parameter to an adaptive parameter that compensates for quantization effects

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If correction coefficient is generated and quantized, then SNR is improved, but code length increases

Engineering Contradiction:

Improvesignal-to-noise ratioVSAvoidcode length

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent performs preliminary quantization of the correction coefficient using a lookup table or codebook before transmission. By pre-defining a finite set of correction values and their corresponding codes, the system achieves compact representation of the correction information, reducing the actual code length while maintaining the SNR improvement benefit

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS8724734B2Coding method, decoding method, apparatuses thereof, programs thereof, and recording medium

Publication Date: 2014.05.13 NIPPON TELEGRAPH & TELEPHONE CORP

AI summary

A coding method with a small error is provided. In the coding method of the present invention, a normalization value obtained from an input signal is corrected for an error calculated from an input and output in vector quantization and is then quantized. The coding method includes a normalization stage of normalizing the input signal in accordance with the normalization value of the input signal, calculated in each frame; a dividing stage of dividing the normalized frame into divided input signal sequences in accordance with a predetermined rule; a vector quantization stage of applying vector quantization to the divided input signal sequences to generate a vector quantization index; and a normalization value correction stage of correcting the normalization value of the input signal for the error obtained from the input and output in the vector quantization stage.