Traffic data classification for security in IoT-based road signaling system

Abstract

Traffic congestion is one of the major problems faced by people in large cities. The traffic controlling systems at present are time stamp oriented which is semiautomatic in nature. With the introduction of IoT in traffic signaling systems, researches are being done considering density as a parameter for automating the traffic signaling system and regulate traffic dynamically. Security is a concern when sensitive data of great volume is being transmitted wirelessly. To prevent the issues on security protocols, we here developed a secured IoT-based system with intelligence that will analyze the traffic data patterns for good or bad and accordingly block attacks to the system. In here, we are addressing man-in-the-middle attack (MITM) only on data for security analysis as a first step. So toward this, support vector machine learning algorithm deployed at the Edge that would classify the traffic data as good or bad based on training before being processed for regulating the traffic signal. The classification of data for applying SVM at the Edge is implemented by taking the raw traffic data set of three cities of Greater London region for 5 years ranging from 2011 to 2016 amounting to 3577. The implementation carried out using Raspberry Pi3 and Scikit.