Hybrid machine learning detection for orbital angular momentum over turbulent MISO wireless channel

Abstract

This work proposes a machine learning detection scheme for wireless orbital angular momentum (OAM) communication systems. The new scheme works for single-input–single-output (SISO) and multiple-input–single-output (MISO) wireless communication between the transmitter and the receiver. The transmitter encodes its data in OAM modes which are constructed using Laguerre–Gaussian beam form. The transmitted beams travel through a wireless channel with weak to medium turbulence strength and they arrive at random positions on the same receiver area. The authors proposed detection scheme allows the reception of multiple overlapping beams without prior knowledge of beams centre positions. The proposed scheme uses a novel technique of receiver segmentation and space filtering along with neural network to decode the received beams. Simulations show the detection efficiency and the enhanced performance of their proposed scheme for the SISO case and the MISO case with 2, 4, and 16 beams.