Averaging Based Predictive Modelling for Traffic Congestion in IoT

The Internet of things (IoT) is the system of physical devices, vehicles, and other items embedded with electronics, software, sensors, actuators, and connectivity which empower these objects to accumulate and interchange data. IoT allows objects to be recognized or controlled distantly without human involvement. This result in enhanced efficiency, precision and economic advantage. Traffic blocking is bursting as foremost challenge in every established as well as emerging countries and it needs immediate attention. The amalgamation of machine learning and IoT, Vehicular Adhoc Network (VANET) makes the traffic management more intelligent. Many researchers have proposed numerous answers for covering detecting, estimating and avoiding traffic congestion in a handful of established nations. These solutions are not suitable from Indian perspective because of mixed traffic conditions, population. This paper proposes a novel traffic congestion prediction technique based on averaging under heterogeneous conditions. The proposed system uses real time and historic traffic data for informing accurately on road congestion preceding the journey.