Energy Estimation for VANET Performance Based Robust Neural Networks Learning

Abstract - Vehicular ad-hoc networks (VANETs) technology has emerged recently as an important research area. They have today been established as reliable networks that vehicles use for communication purpose on highways or urban environments. VANET is a network that is formed when vehicles with wireless transceiver have the need to communicate with each other. It is composed of models based communication among vehicles and vehicle with a high mobility feature. The power consumption by wireless communications might become a major concern in VANET design. Artificial neural networks (ANNs) are one of the most popular and promising areas of artificial intelligence (AI) research. In this paper, we study the performance estimation of VANET in terms of energy consumption and the throughput; we propose the robust neural networks learning that are based on a family of robust statistics estimators, commonly known as M-estimators to replace the traditional MSE performance function order to robustify the neural networks learning in the case of high-quality clean data (noise free). Comparative study between the robust and the traditional performance functions was established in this paper using VANET performance estimation. © 2019, Springer Nature Switzerland AG.