Sebastian Cammerer

TL;DR: This paper builds, train, and run a complete communications system solely composed of NNs using unsynchronized off-the-shelf software-defined radios and open-source deep learning software libraries, and proposes a two-step learning procedure based on the idea of transfer learning that circumvents the challenges of training such a system over actual channels.

TL;DR: The metric normalized validation error (NVE) is introduced in order to further investigate the potential and limitations of deep learning-based decoding with respect to performance and complexity.

TL;DR: In this paper, the authors revisited the idea of using deep neural networks for one-shot decoding of random and structured codes, such as polar codes, and showed that neural networks can learn a form of decoding algorithm, rather than only a simple classifier.

TL;DR: This work partitions the encoding graph into smaller sub-blocks and train them individually, closely approaching maximum a posteriori (MAP) performance per sub-block, and shows the degradation through partitioning and compares the resulting decoder to state-of-the art polar decoders such as successive cancellation list and belief propagation decoding.

TL;DR: This work extends the idea of end-to-end learning of communications systems through deep neural network (NN)-based autoencoders to orthogonal frequency division multiplexing (OFDM) with cyclic prefix (CP) and shows that the proposed scheme can be realized with state-of-the-art deep learning software libraries as transmitter and receiver solely consist of differentiable layers required for gradient-based training.