Perceptron for channel estimation and signal detection in OFDM systems

OFDM (ORTHOGONAL frequency-division multiplexing) is a well-known modulation scheme that has been widely employed in wireless broadband systems in the previous decade to combat frequency-selective type fading in wireless channels. In OFDM approaches, channel state information is critical for detecting and decoding coherent signals. Pilot tones are frequently included into the subcarriers of OFDM signals to perform channel estimation. The perceptron neural network (DNN) has shown to be an effective tool for channel estimation in wireless communication's suboptimal conditions. Prior to the demodulation of OFDM signals, a dynamic channel estimate is important. Depending on the channel types and circumstances, deep learning-based channel estimation outperforms classical channel estimation methods such as minimal mean-square error (MMSE) and least squares (LS). The simulation results validate the projected Perceptron model’s validity and demonstrate the use of our proposed Perceptron-based channel estimation in both nonlinear and linear signal models.