Naive Bayes Classifier Based Driving Habit Prediction Scheme for VANET Stable Clustering

Abstract

Vehicular ad hoc networks (VANETs) is a promising network form for future application on road, like arriving automatic driving and in-vehicle entertainment. Compare with traditional mobile ad hoc networks (MANETs), its advantages are multi-hop communication without energy restriction and relative regular moving pattern. However, the high mobility of nodes raises many challenges for algorithm designers such as topology changing, routing failures, and hidden terminal problem. Clustering is an effective control algorithm provides efficient and stable routes for data dissemination. Efficient clustering algorithms became challenging issues in this kind of distributed networks. In this paper, a novel machine learning based driving habit prediction scheme for stable clustering is proposed, briefly named NBP. In the scheme, vehicles are divided into two alignments with opposite driving habit from which stable cluster design could benefit. Naive Bayes classifier is introduced to estimate the alignment of vehicles by several factors, such as relative speed, vehicle type, number of traffic violations and commercial vehicle or not. Combined with clustering design, the proposed method has been proven effective for stable clustering in VANET.