CSI Partitioning Method with PCA-Based Compression for Low-Complexity Feedback of Large-Dimensional Channels.

This letter considers a codebook-based channel state information (CSI) feedback technique for large-dimensional channels, e.g., massive multiple-input multiple-output (MIMO) channels. We propose a CSI partitioning method, focus on the search complexity required in order to obtain the best beamforming vector, and investigate the tradeoff between the complexity and the performance. Then, we present an analytical framework that provides design guidelines for balancing the complexity, performance, and feedback overhead of practical large-scale MIMO systems.