Human–Machine Interaction Using Probabilistic Neural Network for Light Communication Systems

Hand gestures are a natural and efficient means to control systems one of the promising but challenging areas human–machine interaction (HMI). We propose system recognize by processing interrupted patterns light in visible communications (VLC) system. Our solution is aimed at emerging communication can facilitate human–computer for services health-care, robot systems, commerce home. The exploits existing infrastructure using low-cost readily available components. Different finger sequences detected probabilistic neural network (PNN) trained on transitions between fingers. A novel pre-processing sampled photodiode described use PNN with limited complexity. contributions this work include development sensing technique methodology convert into manageable size matrices along hardware implementation showing proof concept under lighting conditions. Despite modest complexity our could correctly an accuracy 73%, demonstrating potential technology. show that depends matrix Gaussian spread function. IEEE 802.11bb ‘Li-Fi’ standard expected bring virtually every room across world exploit gesture be considerable interest value society.