Universal Unitary Space Vector Quantization Codebook Design for Precoding MIMO System under Spatial Correlated Channel
Ping Wu, Lihua Li, Ping Zhang
Abstract
Ping Wu, Lihua Li, Ping Zhang
Abstract
In codebook based precoding MIMO system, the precoding codebook affects the system performance remarkably. Consequently, it is a crucial technology to design the optimum codebook for precoding MIMO system. Codebook design is related to the channel fading, antenna number, spatial correlation etc. So specific distributed channel corresponds to respective optimum codebook. In this paper, in order to design the optimum codebooks for the precoding MIMO system, a universal unitary space vector quantization (USVQ) codebook design criterion is provided, which can design the optimum codebooks for various spatial correlated channels with arbitrary antenna configurations. The unitary space K-mean(USK) algorithm is also provided to generate the USVQ codebook, which is iterative and convergent. Simulations show that the capacities of the precoding MIMO schemes using the USVQ codebooks are very close to those of the ideal precoding cases and outperform those of the schemes using the traditional Grassmannian codebooks.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In codebook based precoding MIMO system, the precoding codebook affects the system performance remarkably. Consequently, it is a crucial technology to design the optimum codebook for precoding MIMO system. Codebook design is related to the channel fading, antenna number, spatial correlation etc. So specific distributed channel corresponds to respective optimum codebook. In this paper, in order to design the optimum codebooks for the precoding MIMO system, a universal unitary space vector quantization (USVQ) codebook design criterion is provided, which can design the optimum codebooks for various spatial correlated channels with arbitrary antenna configurations. The unitary space K-mean(USK) algorithm is also provided to generate the USVQ codebook, which is iterative and convergent. Simulations show that the capacities of the precoding MIMO schemes using the USVQ codebooks are very close to those of the ideal precoding cases and outperform those of the schemes using the traditional Grassmannian codebooks.
Key concepts: Codebook, Precoding, MIMO, Linde–Buzo–Gray algorithm, Algorithm, Zero-forcing precoding, Spatial correlation, Fading