Экспериментальное Исследование Самоорганизующегося Акустического Канала Связи

Abstract

This paper presents the results of an experimental study of the adaptive communication system with phase separation. Adaptation is carried out by methods of machine learning and consists in transmitters phases regulation. An acoustic waveguide with a tunable metasurface is used as a propagation medium. In the case of multi-beam propagation and the presence of significant inter-symbol interference the transmission rate of 1 kbit/s was achieved. The adaptation time of the investigated system to sudden changes in the channel parameters does not exceed 10 cycles of training, even in case of significant fading.