Reproduce “OFDM-Autoencoder for End-to-End Learning of Communications Systems” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Alexander Felix, Sebastian Cammerer, Sebastian Dörner, Jakob Hoydis, Stephan ten Brink
The authors integrate trained neural transmitters and receivers into an OFDM system with cyclic prefixes. Messages occupy subcarriers, while the prefix supports synchronization and a Fourier transform separates them at reception. In simulations, the system achieves an approximately 2 dB advantage in block error rate over QPSK with MMSE equalization at the same spectral efficiency; the evidence is simulated, not an over-the-air demonstration.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.