Abstract

Downlink non-orthogonal multiple access (NOMA) is necessary to handle the underwater acoustic channel's limitations on bandwidth. NOMA downlink underwater acoustic (UWA) communication transmits data symbols from a source station to many users. Superimposed coding with variable power levels allows successive interference cancelation (SIC) receivers to decode data. However, SIC receivers require knowledge of channel conditions and channel state information (CSI). This is difficult to acquire, particularly in UWA communication. To address this problem, this paper proposes downlink underwater acoustic using 1D Convolution neural network (CNN). Two users with different power levels and distances from the transmitter employ BPSK modulation to support multiuser communication. Users far from the base station receive the most power. The base station uses superimposed coding. BELLHOP algorithm generates the training dataset with user depth modifications. For training the model, a composite signal passes through the samples of the UWA channel and is fed to the model along with labels. DNN receiver learns the characteristic of the UWA channel and does not depend on CSI. The testing CIR is used to evaluate the trained model. The results are compared to the traditional SIC receiver. The DNN-based DL NOMA underwater acoustic receiver outperformed the SIC receiver in simulations.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call