2008Unpublished venueRequires access

Precoding Vector Distribution under Spatial Correlated Channel and Nonuniform Codebook Design

H. Long, Weidong Wang, Hui Zhao, Kan Zheng

Open publisher page 7 citations

Abstract

In this paper, influence of spatial correlation on codebook-based precoding technology is analyzed. Optimum pre- coder distribution under spatial correlated channel is achieved at first, which is no longer isotropical on the unitary matrix (vector) set. Selection probability of the matrix (vector) in codebook is not uniform too. Then, a novel codebook design is presented for spatial correlated channel, which is termed as nonuniform codebook. Codebook is designed as function of spatial correlation characteristic of channel and presents a nonuniform quantization of unitary vector set. Smaller quantization error and higher spectral efficiency can be achieved by proposed codebook design than that of traditional fixed uniform codebook design, with the cost of slight feedback overhead increasing.

About this research paper

What this paper is about

In this paper, influence of spatial correlation on codebook-based precoding technology is analyzed. Optimum pre- coder distribution under spatial correlated channel is achieved at first, which is no longer isotropical on the unitary matrix (vector) set. Selection probability of the matrix (vector) in codebook is not uniform too. Then, a novel codebook design is presented for spatial correlated channel, which is termed as nonuniform codebook. Codebook is designed as function of spatial correlation characteristic of channel and presents a nonuniform quantization of unitary vector set. Smaller quantization error and higher spectral efficiency can be achieved by proposed codebook design than that of traditional fixed uniform codebook design, with the cost of slight feedback overhead increasing.

Why it matters

OpenAlex reports 7 citations for this work. Citation counts describe recorded attention and do not establish research quality.

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

In this paper, influence of spatial correlation on codebook-based precoding technology is analyzed. Optimum pre- coder distribution under spatial correlated channel is achieved at first, which is no longer isotropical on the unitary matrix (vector) set. Selection probability of the matrix (vector) in codebook is not uniform too. Then, a novel codebook design is presented for spatial correlated channel, which is termed as nonuniform codebook. Codebook is designed as function of spatial correlation characteristic of channel and presents a nonuniform quantization of unitary vector set. Smaller quantization error and higher spectral efficiency can be achieved by proposed codebook design than that of traditional fixed uniform codebook design, with the cost of slight feedback overhead increasing.

Key concepts: Codebook, Linde–Buzo–Gray algorithm, Precoding, Algorithm, Vector quantization, Spatial correlation, Computer science, Overhead (engineering)

Related papers

Back to paper searchBrowse research topicsOriginal source
Precoding Vector Distribution under Spatial Correlated Channel and Nonuniform Codebook Design — Research Paper | ScholarLens