Deep Learning for 2D-MIMO scheme based on Optical Camera Communication

Abstract

Radio frequency (RF) waveforms are widely used in wireless technologies, such as mobile system, smart factory, and smart home systems. Due to its simplicity of installation, wireless communication is better than wire communication. However, the use of high frequencies for data transmission can be harmful to human health. To replace RF waveform for communication system, Optical Wireless Communication (OWC), which deploys visible light waveform for communication, was researched widely by researchers around the world. Visible Light Communication (VLC), Light Fidelity (LiFi), and Optical Camera Communication (OCC) are the three candidates for replacing RF system. In this research, we present an multiple-input multiple-output (MIMO) modulation system based on on-off keying (OOK) modulation in the time domain and a light-emitting diode (LED) array. By adjusting the exposure time, focal length, forward error correction, and Deep learning for object detection, it is still possible to implement this technique with 10 links in various locations at a distance of 20 m with a velocity of 2 m/s.