Dr Muhammad Hilmi Kamarudin

Connected and Autonomous Vehicles (CAVs) rely on Vehicular Adhoc Networks with wireless communication between vehicles and roadside infrastructure to support safe operation. However, cybersecurity attacks pose a threat to VANETs and the safe operation of CAVs. This study proposes the use of simulation for modelling typical communication scenarios which may be subject to malicious attacks. The Eclipse MOSAIC simulation framework is used to model two typical road scenarios, including messaging between the vehicles and infrastructure - and both replay and bogus information cybersecurity attacks are introduced. The model demonstrates the impact of these attacks, and provides an open dataset to inform the development of machine learning algorithms to provide anomaly detection and mitigation solutions for enhancing secure communications and safe deployment of CAVs on the road.

With the rapid growth of the Internet, there are an increasing number of computer threats and attacks. The prevalence of zero-day attack activities has given rise to the need to prevent these attack activities from spreading and damaging the computer system. As such, intrusion detection system (IDS) should satisfy complex requirements and must be durable, manageable and reliable. In this paper, we developed an anomaly-based detection model using a statistical method combined with a binary logistic regression approach. The model, Layer based Anomaly Detection (LbAD) is designed to detect remote to user (R2L) and user to root (U2R) attacks by statistically examining the degree of normal field values within three layer (data link, network, transport) of OSI Seven Layer. The results of the new method outperform the leading existing methods.

Organization has come to realize that network security technology has become very important in protecting its information. With tremendous growth of internet, attack cases are increasing each day along with the modern attack method. One of the solutions to this problem is by using Intrusion Detection System (IDS). Machine Learning is one of the methods used in the IDS. In recent years, Machine Learning Intrusion Detection system has been giving high accuracy and good detection on novel attacks. In this paper the performance of a Machine Learning algorithm called Decision Tree (J48) is evaluated and compared with two other Machine Learning algorithms namely Neural Network and Support Vector Machines which has been conducted by A. Osareh [1] for detecting intrusion. The algorithms were tested based on accuracy, detection rate, false alarm rate and accuracy of four categories of attacks. From the experiments conducted, it was found that the Decision tree (J48) algorithm outperformed the other two algorithms.

The rapid growth in the volume and importance of web communication throughout the Internet has heightened the need for better security protection. Security experts, when protecting systems, maintain a database featuring signatures of a large number of attacks to assist with attack detection. However used in isolation, this can limit the capability of the system as it is only able to recognize known attacks. To overcome the problem, we propose an anomaly-based intrusion detection system using an ensemble classification approach to detect unknown attacks on web servers. The process involves removing irrelevant and redundant features utilising a filter and wrapper selection procedure. Logitboost is then employed together with random forests as a weak classifier. The proposed ensemble technique was evaluated with some artificial data sets namely NSL-KDD, an improved version of the old KDD Cup from 1999, and the recently published UNSW-NB15 data set. The experimental results show that our approach demonstrates superiority, in terms of accuracy and detection rate over the traditional approaches, whilst preserving low false rejection rates.