Implementation of deep learning based method for optimizing spatial diversity mimo communication

As an alternative solution of the isuue trade-off phenomenon between performance and computational complexity always become the hugest dilemma suffered by researchers, this research proposes an optimization in spatial diversity MIMO communication system using end-to-end learning based model, specifically, it adapts autoencoder model. Two models are introduced in this research which each of them address a problem about data detection task and channel estimation task that has not been addressed in the previous research. The proposed models were evaluated in one of the most common channel impairment which is Rayleigh fading with additional Additive White Gaussian Noise (AWGN) and compared to the standard Alamouti scheme. The results show that these deep learning based models for MIMO communication system result in very promising results by outperforming the baseline methods. In perfect CSIR (Channel State Information in Receiver side) case, the proposed models achieve BER nearly 10⁻⁵ at SNR 22.5 dB. While in channel estimation case, the proposed models can exceed the baseline performance even by only transmitting 2 pilots.