Visual-MIMO for vehicle to vehicle communications

In most of the scenarios, accidents occur due to the carelessness of the driver. To reduce the accidents sharing of the vehicle's dynamics information is essential. The primary vehicle uses several sensors to collect information which were transmitted to the next subordinate vehicles that follow it. For communicating huge data between vehicles we need a Multiple Input Multiple Output Radio Frequency (MIMO-RF) transceiver which leads to increase in complexity and cost of vehicle design. This drawback can be overcome by using the existing Visible Light Communication (VLC), because of the accessibility of light source and probability of seamless communication. In this paper, a Light Emitting Array (LEA) setup is likely to be placed which is the Region of Interest (ROI). The comparative distance between the camera and the LEA varies dynamically that leads to increase in average Bit Error Rate (BER) at the output of the detector. To overcome this problem, we proposed the Kalman filter methodology along with regular geometry techniques to locate and track the LEA and the deformed shape of LEA is corrected utilizing perspective projection that maximizes the average BER performance. The simulation results show that the proposed scheme has an optimum performance than the existing method.