ScholarLens Back to search
2008•Microelectronics & ComputerRequires access

Codebook Design Algorithm for Vector Quantization Based on Particle Swarm Optimization

Cai Guang-yue

Open publisher page 0 citations

Abstract

A codebook design algorithm for vector quantization based on particle swarm optimization(PSO)is proposed. To solve the shortcoming that the initial codebook selection affects the convergence speed of codebook training and the per- formance of the final codebook,the initial codebook with global characteristics is produced using PSO,and iteratively pro- duce final codebook with local characteristics using Linde-Buzo-Gray(LBG).Both of the theory and practical effect show the rationality of the proposed algorithm.

About this research paper

What this paper is about

A codebook design algorithm for vector quantization based on particle swarm optimization(PSO)is proposed. To solve the shortcoming that the initial codebook selection affects the convergence speed of codebook training and the per- formance of the final codebook,the initial codebook with global characteristics is produced using PSO,and iteratively pro- duce final codebook with local characteristics using Linde-Buzo-Gray(LBG).Both of the theory and practical effect show the rationality of the proposed algorithm.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

A codebook design algorithm for vector quantization based on particle swarm optimization(PSO)is proposed. To solve the shortcoming that the initial codebook selection affects the convergence speed of codebook training and the per- formance of the final codebook,the initial codebook with global characteristics is produced using PSO,and iteratively pro- duce final codebook with local characteristics using Linde-Buzo-Gray(LBG).Both of the theory and practical effect show the rationality of the proposed algorithm.

Key concepts: Codebook, Linde–Buzo–Gray algorithm, Vector quantization, Computer science, Particle swarm optimization, Algorithm, Convergence (economics), Quantization (signal processing)

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Codebook Design Algorithm for Vector Quantization Based on Particle Swarm Optimization — Research Paper | ScholarLens