Novel Codebook Design and Performance Analysis for Time-Varying Spatial Channels With FDD MIMO

Abstract

Codebook design utilizing channel statistics is an effective solution to reduce feedback overhead in frequency-division duplexing (FDD) multiple-input multiple-output (MIMO) systems. In this paper, we propose a novel codebook designed for time-varying spatial channels and analyze its quantization error (QErr). To this end, we first develop a new analysis on the average QErr of the channel statistics-rotated (CSR) codebook in a time-invariant spatial channel. The developed analysis is expressed as a function of the eigenvalue profiles of the channel statistics matrix, even when its rank remains constant. Next, by leveraging our newly developed analysis for the CSR codebook, we propose a novel time-varying adaptive CSR (TACSR) codebook in a time-varying spatial channel and derive the average QErr to characterize its performance from the channel statistics. Simulation results show that our new analysis for the CSR codebook provides much tighter bounds, while the proposed TACSR codebook significantly outperforms the existing codebooks in time-varying spatial channels.