Limited Feedback for X-Structured Precoding in Spatial-Multiplexing MIMO Systems

Abstract

X-structured precoding has been considered a promising precoding method for MIMO systems with maximum-likelihood (ML) receivers. It not only simplifies the precoder design, but allows a low-complexity ML detection scheme. For a real-world precoding scheme, a precoder can only be selected from a codebook and its index is fed back to the transmitter, known as limited-feedback precoding. Due to its special structure, two codebooks are required in X-structured precoding. In this paper, we first investigate how to construct the codebooks for X-structured precoding. Then, we consider the detection problem in limited-feedback X-structured precoding in which the X-structure no longer holds. With group successive interference cancelation, we propose a MIMO detection method to solve the problem. The synergy of the proposed methods yields a low-complexity ML precoding and detection scheme. Simulations show that the proposed methods provide satisfactory results.