Deep Learning for Channel Estimation and Signal Detection in OFDM-Based Communication Systems

Abstract

The goal of 6G communication networks requires higher transmission speeds, tremendous data processing, and low-latency communication. Orthogonal frequency-division multiplexing (OFDM), which is widely utilized in 5G communication systems, may be a viable alternative for 6G. It significantly reduces inter symbol interference (ISI) in the frequency-selective fading environment. Channel estimation is critical in OFDM to optimize system performance. Deep learning has been employed as an appealing alternative for channel estimation and signal detection in OFDM-based communication systems due to its better potential for feature learning and representation. In this study, we examine the deep neural network (DNN) layers created from long-short term memory (LSTM) for detecting the signals by learning the received signal as well as channel information. We investigate the performance of the system under various conditions. The simulation results show that the signal bit error (SER) is equivalent to and better than that of the minimum mean squared error (MMSE) and least square (LS) methods.