Tensor-Based Signal Processing with Applications to MIMO-OFDM Systems and Intelligent Reflecting Surfaces

Abstract

The use of tensor algebra techniques in signal processing has been growing over the last two decades. Applications like image processing, biomedical signal processing, radar, machine/deep learning, and communications in general, largely employ tensor-based techniques for recovery, estimating, and classifying signals. One of the main reasons for using tensor signal processing is the exploitation of the multidimensional structure of signals, while benefiting from the uniqueness properties of tensor decomposition. For example, in wireless communications, the signals can have several ”dimensions", e.g., space, time, frequency, polarization, beamspace, etc.. This thesis is divided into two parts, first, in the application of a tensor-based algorithm in multiple-input multiple-output (MIMO)-orthogonal frequency division multiplexing (OFDM) systems with the presence of phase-noise impairments. In this first part, we propose a two-stage tensor-based receiver for a joint channel, phase-noise, and data estimation in MIMO-OFDM systems. In the first stage, we show that the received signal at the pilot subcarriers can be modeled as a third-order PARAFAC tensor. Based on this model, we propose two algorithms for channel and phase-noise estimation at the pilot subcarriers. The second stage consists of data estimation, for which we propose a ZF receiver that capitalizes on the tensor structure of the received signal at the data subcarriers using the proposed SKP operator. Numerical simulations show that the proposed receivers achieves an improved performance compared to the state-of-art receivers in terms of symbol error rate (SER) and normalized mean square error (NMSE) of the estimated channel and phase-noise matrices. The second part of this thesis focuses on the application of tensor modeling to reduce the control signaling overhead in future wireless systems aided by intelligent reflecting surfaces (IRS). To this end, we propose a low-rank approximation of the near-optimal IRS phase-shifts, which would incur prohibitively high communication overhead on the BS-IRS controller links. The key idea is to represent the potentially large IRS phase-shift vector using a low-rank tensor model. This is achieved by factorizing a tensorized version of the IRS phase-shift vector, where each component is modeled as the Kronecker product of a predefined number of factors of smaller sizes, which can be obtained via tensor decomposition algorithms. We show that the proposed low-rank models drastically reduce the required feedback requirements associated with the BS-IRS control links. Simulation results indicate that the proposed method is especially attractive in scenarios with a strong line of sight component, in which case nearly the same spectral efficiency is reached as in the cases with near-optimal phase-shifts, but with a drastically reduced communication overhead.