A Design of the End-to-End Wireless Communication System Based on Autoencoder and ADALM-PLUTO

The application of deep learning in the field of wireless communication is a research hotspot in current academia. Especially, the research of end-to-end wireless communication technology based on autoencoder is gradually carried out, but there are few engineering cases. Therefore, this paper proposes a design scheme of end-to-end wireless communication system based on autoencoder and ADALM-PLUTO, which can be used for prototype verification and wireless communication. Firstly, a wireless communication system was constructed by using computers, ADALM-PLUTO SDR platforms and wireless communication neural network trained based on autoencoder. Secondly, the SDR software kernels of signal transmitting and receiving and the system operation processes were explained in detail. Finally, the training data was constructed and the experiments were carried out. The results show that the proposed scheme can achieve better end-to-end wireless communication in indoor environment, and has certain reference value for engineering implementation and technical application.