Smart System for Detection of BOTs in Cyber Physical Environment

Abstract

Botnets has become the serious threat against cyber security for the cyber physical devices. For analyzing and investigating such attacks the important way is to observe the botnet network traffic. Botnet attacks classified as topology based, protocol based, architecture based. Designing a detection system for bots is becoming challenging as botnet attacks are upgrading the attacking methodology (Architecture, protocol, topology) periodically. The main aim of this paper is to investigate various bot detection algorithms and their architecture. Moreover, paper also focuses on application based data and network based data. The analysis is based on type of botnet attack, detection target, feature source, feature extraction, feature correlation, machine learning techniques. As a result, this paper is proposing an architecture, protocol, topology independent network-based early alert based system. The proposed model is analyzing the network traffic, and based on various correlation, classification techniques generates alert for presence of bot in the network.