The proliferation of Internet of Things (IoT) comes with it scores of end devices without proper security design and this adds to the challenge of network security. Also, the smartphones and computers use countless apps and applications for online purchases, and the user needs to use it securely. With many wired and wireless smart devices available, to perform network attacks like spoofing, flooding, eavesdropping, masquerading etc. is so easy, as so many open source hack tools are available. So, the rationale for this research is to develop an intelligent intrusion detection system (IDS) which will detect many such attacks.
A normal IDS looks for different kind of network attacks in the incoming packets. This project will use machine learning (ML) to differentiate the attack packet flows and classify the attacks, which will allow the learning algorithm to understand new kind of attacks and grow in intelligence. This study will use a subset of machine learning called deep learning (DL) for intrusion detection using Recurrent Neural Networks (RNN) and other DL variants. RNN is very suitable for modelling the classification with high accuracy and its performance is superior to that of traditional ML classification methods in both binary and multiclass classification. Variants of RNN like Neural history compressor, Second order RNNs, Long short-term memory, Gated recurrent unit, Bi-directional, Continuous-time, Hierarchical, Recurrent multilayer perceptron network, Multiple timescales model etc. will be studied. Optimizing the deep learning algorithms will be the next step through Stochastic Gradient Descent methods (SGDs), Limited memory BFGS (L-BFGS) and Conjugate Gradient (CG) with line search etc. or with nature-inspired meta-heuristic optimization algorithms like Ant Colony Optimization, Particle Swarm Optimization, Firefly Algorithm, Honey Bee Algorithm, Whale Optimization Algorithm etc. We will compare our work with the state-of-the-art and will develop a superior architecture for intrusion detection.
The principal supervisor for this project is Biju Issac.
Eligibility and How to Apply:
Please note eligibility requirement:
• Academic excellence of the proposed student i.e. 2:1 (or equivalent GPA from non-UK universities [preference for 1st class honours]); or a Masters (preference for Merit or above); or APEL evidence of substantial practitioner achievement.
• Appropriate IELTS score, if required.
• Applicants cannot apply for this funding if currently engaged in Doctoral study at Northumbria or elsewhere.
In addition to the aforementioned eligibility criteria, the applicants should have good skills in the following areas: Network Security, Machine Learning and Artificial Intelligence algorithms. The following skills are highly desirable for this project:
• Ability to read and understand state of the art research in network security and machine/deep learning
• Experience in Matlab
• Computer programming using object-oriented language (preferably Java, Python or C++)
• Analytical modelling
• Application development experience
• A good conference or journal publication in the related areas
For further details of how to apply, entry requirements and the application form, see https://www.northumbria.ac.uk/research/postgraduate-research-degrees/how-to-apply/
Please note: Applications that do not include a research proposal of approximately 1,000 words (not a copy of the advert), or that do not include the advert reference (e.g. RDF19/EE/CIS/ISSAC) will not be considered.
Deadline for applications: Friday 25 January 2019
Start Date: 1 October 2019
Northumbria University is an equal opportunities provider and in welcoming applications for studentships from all sectors of the community we strongly encourage applications from women and under-represented groups.
Shah, S. A. R., & Issac, B. (2018). Performance Comparison of Intrusion Detection Systems and Application of Machine Learning to Snort System, Future Generation Computer Systems, Elsevier, ISSN 0167-739X, Vol. 80, 157-170.
Sharma, R. K., Kalita, H. R. & Issac, B. (2018). Are Machine Learning Based Intrusion Detection System Always Secure? An Insight into Tampered Learning, Journal of Intelligent and Fuzzy Systems, IOS Press, ISSN 1064-1246, 17 pages.
Shah, S. A. R., Issac, B. & Jacob, S. M. (2018). Intelligent Intrusion Detection System through Combined and Optimized Machine Learning, International Journal of Computational Intelligence and Applications (IJCIA), Imperial College Press (World Scientific Europe), ISSN 1469-0268, 17(2), 17 pages.
Albayati, M., & Issac, B. (2015). Analysis of Intelligent Classifiers and Enhancing the Detection Accuracy for Intrusion Detection System, International Journal of Computational Intelligence Systems (IJCIS), Atlantis Press, Taylor and Francis, ISSN 1875-6891, 8(5), 841-853.
Sharma, R. K., Kalita, H. R. & Issac, B. (2016). PIRIDS: A Model on Intrusion Response System Based on Biologically Inspired Response Mechanism in Plants, in "Advances in Intelligent and Soft Computing" Eds. Snášel V., Abraham, A., Krömer, P., Pant, M. & Muda, A. K., Springer, ISBN 978-3-319-28031-8, pp.105-116
Barlaskar, E., Singh, J. Y., & Issac, B. (2016). Energy-Efficient Virtual Machine Placement using Enhanced Firefly Algorithm, Multiagent and Grid Systems, IOS Press, ISSN 1574-1702, 12(3), 167-198.
Barlaskar, E., Singh, J. Y., & Issac, B. (2018). Enhanced Cuckoo Search Algorithm for Virtual Machine Placement in Cloud Data Centers, International Journal of Grid and Utility Computing, Inderscience, ISSN 1741-8488, 9(1), 1-17.