Mobile Crowdsensing based Architecture for Intelligent Traffic Prediction and Quickest Path Selection

Abstract

The mobile crowd sensing (MCS) network is a new reliable and robust paradigm It consists of the Internet of Things (IoTs), wireless sensor networks (WSNs), and mobile personal devices. MCS is commonly used for social, infrastructural, and environmental data collection. Therefore, MCS architecture is utilized for a real-time traffic flow measurement and for predicting the quickest path. Nowadays, traffic congestion is becoming a severe concern in urban areas. The main reasons for traffic congestion and traffic jam on the roads of metropolitan cities are ever-increasing population and vehicle production. Therefore, in this paper, we propose an MCS system, which provides the user congestion-free path to reach the destination in minimum travel time. The MCS architecture exploits collected smartphone data (e.g., speed, direction, and location) for a real-time traffic prediction. Subsequently, K-means Clustering is used to divide the traffic into small clusters. Then the convex hull algorithm is used to calculate the weights of each cluster. In this manner, the proposed system can competently determine the quickest path. The MCS system updates the user about the real-time traffic flow and suggests the quickest path after a specific interval of time until the user reaches the destination. We evaluate the proposed system by comparison with the traditional systems. The obtained results demonstrate that the proposed system provides less distance and reduces the travel time for different traffic scenarios as compared to traditional systems.