ORTHOGONAL FREQUENCY-DIVISION MULTIPLEXING EQUALIZATION USING DEEP NEURAL NETWORK

Orthogonal frequency-division multiplexing (OFDM) equalization using a Deep Neural Network (DNN) may be provided. First, a signal in a packet structure may be received at an OFDM receiver from an OFDM transmitter. The signal may have distortion. Training constellation points, pilot constellation points, and data constellation points may be extracted from the signal based on the packet structure. Each data constellation point may correspond to a data subcarrier within a data symbol of the signal. Next, the training constellation points and the pilot constellation may be provided as input for the data symbol to a DNN. A coefficient for each data subcarrier within the data symbol that reverses the distortion may be received as output from the DNN. Then, the coefficient for each data subcarrier may be applied to the corresponding data constellation point to determine a per subcarrier constellation point prediction.