Multiantenna Multicarrier Transceiver Architectures

Abstract

In wireless communications, increased spectral efficiency and low error rates can be achieved by means of space-time-frequency coded MIMO-OFDM systems. In this work, we illustrate tensor-based approaches for MIMO-OFDM systems combining space-frequency and time domain processing allowing iterative joint blind channel estimation and symbol decoding. First, we consider the case of MLSFC with an extended LCP. Then, space-time-frequency signaling technique that combines space-frequency modulation with a time-varying linear precoding is developed. We show that both systems satisfy PARAllel FACtor (PARAFAC)-based models, which allow a blind joint channel and symbol estimation using iterative or closed-form receiver algorithms. For this system, we propose two closed-form semi-blind receivers that exploit differently the multilinear structure of the received signal, which is formulated as a nested PARAFAC model. For the first system, ALS and LS-KRF receivers are proposed and compared. For the later system, and aiming at reducing pilot overhead, we develop a S-CFP receiver coupled with a pairing algorithm that yields an unambiguous estimation of the transmitted symbols without the need of a pilot frame. Simulation results are shown to evaluate the performance of the proposed transceivers in terms of bit error rate and channel estimation accuracy.