Non-line-of-sight target tracking with improved recurrent extreme learning machine

Abstract Target tracking provides important location-based services in many applications. The main challenge of target is to combat the severe degradation problem Non-Line-of-Sight (NLOS) scenario. Most Deep Learning algorithms available literature address this issue belong batch learning with high complexity. This paper proposes a novel online sequential algorithm, Improved Recurrent Extreme Machine (IRELM), solve NLOS as position series prediction task. IRELM able train Neural Network (RNN) inputs one-by-one and adaptively update input weight, hidden feedback weight output weight. Extensive simulations experiments prove superior performance feasible complexity over state-of-the-art methods.