2021/01/20 by Fayçal Ait Aoudia, Jakob Hoydis, Aoudia, Fayçal Ait +1 · 1 citation
Computer Science · Engineering · #Advanced Wireless Communication Techniques #FOS: Computer and information sciences #FOS: Electrical engineering #Information Theory (cs.IT) #Machine Learning (cs.LG) #Signal Processing (eess.SP) #Speech and Audio Processing #Wireless Signal Modulation Classification #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2101.08213
openalex publication_date 2021/01/20 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Orthogonal frequency division multiplexing (OFDM) is one of the dominant\nwaveforms in wireless communication systems due to its efficient\nimplementation. However, it suffers from a loss of spectral efficiency as it\nrequires a cyclic prefix (CP) to mitigate inter-symbol interference (ISI) and\npilots to estimate the channel. We propose in this work to address these\ndrawbacks by learning a neural network (NN)-based receiver jointly with a\nconstellation geometry and bit labeling at the transmitter, that allows CP-less\nand pilotless communication on top of OFDM without a significant loss in bit\nerror rate (BER). Our approach enables at least 18% throughput gains compared\nto a pilot and CP-based baseline, and at least 4% gains compared to a system\nthat uses a neural receiver with pilots but no CP.\n