Experimental multiuser mobile optical communication using compressive sensing

Abstract

This paper presents a multiuser mobile optical wireless communication system that uses a single digital camera to receive signals transmitted by multiple users using light-emitting diode (LED) arrays. The key idea is to sample only the engaged camera pixels affected by the signals from the users. Toward this aim, image frames received during an acquisition phase are used to estimate the signal support of the users in the image plane using the orthogonal matching pursuit (OMP) algorithm in a compressed sensing framework. The detected signal support is then clustered and tracked using Kalman Filters. Both numerical methods and hardware experiments are used to validate the algorithms. The proposed system is found to provide excellent performance even for users with weak transmitted signals.