Signal Detection of Mobile Multi-user Molecular Communication System Using Transformer-Based Model

This paper proposes a signal detector for mobile multi-user molecular communication system by using Transformer-based model. This detector can improve the accuracy of detection ability by training the Transformer-based model at different initial distances between the transmitters and receiver, and can also perform detection in the case of unknown channel parameters. Numerical results show that compared with deep neural networks model, the Transformer-based model performs better detection ability in mobile multi-user molecular communication system with lower bit error rate of signal detection.