Recent research in the design of end to end communication system using deep learning has produced models which can outperform traditional communication schemes. Most of these architectures leveraged autoencoders to design the encoder at the transmitter and decoder at the receiver and train them jointly by modeling transmit symbols using encoders as latent codes. However, in communication systems, the receiver has to work with noise corrupted versions of transmit symbols. Traditional autoencoders are not designed to work with latent codes corrupted with noise. In this work, we provide a framework to design end to end communication systems which accounts for the existence of noise corrupted transmitted symbols. The proposed method uses deep neural architecture and an objective function for optimizing these models is derived based on the concepts of variational inference. Further, domain knowledge such as channel type can be easily systematically integrated into the objective and we provide examples on how to do this in the cases of AWGN and RBF channels. Through experimental validation, the proposed method is shown to produce better models consistently in terms of error rate performance as well as constellation packing density as compared to previous works leveraging deep learning methods.
View on arXiv