Neural Network Based Data Estimation for Unique Word OFDM
Sprache des Vortragstitels:
Asilomar Conference on Signals, Systems, and Computers (ACSSC 2021)
Sprache des Tagungstitel:
Model-based methods have been employed for data estimation, also termed equalization, for several decades. Due to the incredible success of data-driven machine learning methods, efforts have recently been made to utilize neural networks (NNs) for data estimation in general multiple-input multiple-output (MIMO) communication systems. In this work, NN based data estimators are investigated for a communication system employing the unique word orthogonal frequency division multiplexing (UW-OFDM) signaling scheme. We evaluate their bit error ratio performance in indoor frequency selective environments, we discuss the pros and cons of the individual approaches, and we highlight their partially peculiar estimation characteristics.