Deep Learning Based Communication Over the Air. IEEE Journal of Selected Topics in Signal Processing, (12)1:132-143, February 2018. [PUMA: myown network;over-the-air;software-defined software-defined from:sdnr transmissions;open-source (artificial systems;block-based module;transmitter transmission;receiver implementations;deep computing;two-step nets;radio intelligence);neural libraries;continuous learning;communications learning;end-to-end rate;over-the-air synchronization;frame error transmissions;receiver deep neural software synchronization data learning;modulation;neural learning libraries;software radio implementations;off-the-shelf networks;NNs;block procedure;end-to-end radio;synchronisation;telecommunication radios;Training;Receivers;Communication networks;Hardware;Transmitters;Synchronization;Autoencoder;communication;deep receivers;software systems;Artificial]
OFDM-Autoencoder for End-to-End Learning of Communications Systems. 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), 1-5, June 2018. [PUMA: amplifier;frequency-selective myown from:sdnr network-based autoencoders;single-tap computing;end-to-end (artificial networks;Synchronization;Training;Signal processing;Robustness libraries;synchronisation;telecommunication channels;cyclic fading modulation;software training;deep oscillators;commodity multiplexing;OFDM;Receivers;Artificial errors;single-carrier prefix;orthogonal implementations;reliable libraries;imprecise equalisers;fading hardware;multipath intelligence);multipath neural software synchronization frequency communication;OFDM-autoencoder;nonlinear learning equalization;sampling channels;gradient-based channels;learning division learning;deep channels;neural nets;OFDM]