Receptores MIMO baseados em algoritmo de decomposição PARAFAC

Abstract

This work deals with the application of multi-way analysis to the context of signal processing for wireless communications. A tensor decomposition known as PARAFAC (PARAllel FACtors) is considered in the design of multiple-input multiple-output (MIMO) receiver for a wireless communication system with Spread Spectrum codes. We propose two supervised PARAFAC-based receiver structures for joint symbol and channel estimation. The first one is based on the knowledge of the spreading codes and the second on the knowledge of a training sequence per transmit antenna. Two code structures are considered, which are Hadamard-Wash (HW) and Truncated Hadamard-Walsh (THW). The main advantages of the proposed PARAFAC receivers is on the fact that they exhibit fast convergence and eliminate all ambiguities inherent to the PARAFAC model. Simulation results are provided to compare the performances of the two receivers for several systems configurations, revealing the impact of the number of transmit antennas, number of receiver antennas, code length and signal to noise ratio in their performances.