Smartphone camera-based device-to-device communication using neural network-assisted high-density modulation

Abstract

Device-to-device (D2D) communication has become one of the focuses for future wireless communications in conjunction with rapidly growing smartphone technologies. Smartphone camera-based D2D communication is presented as a favorable solution to supporting indoor D2D communications. As a smartphone has both camera and display, two smartphones can communicate over a relatively short range distance indoors in the form of smartphone display (transmitter) and smartphone camera (receiver) link. The smartphone camera-based wireless communication called optical camera communication (OCC), however, suffers from low data rate that is at most a hundred kilobits per second. To ameliorate this, we propose a high-density modulation that is more efficient in achieving an Mbps rate for smartphone camera-based D2D systems, by exploiting two-dimensional data transmission capability in OCC. Neural network is also implemented to improve the performance of the proposed system. To verify the proposed system, simulations and experiments were performed. It is demonstrated that a 2.66 Mbps rate of the proposed system is achieved with an acceptable bit error rate.