Visible to the public Biblio

Filters: Keyword is streaming analytics  [Clear All Filters]
Ferrando, Roman, Stacey, Paul.  2017.  Classification of Device Behaviour in Internet of Things Infrastructures: Towards Distinguishing the Abnormal from Security Threats. Proceedings of the 1st International Conference on Internet of Things and Machine Learning. :57:1–57:7.

Increasingly Internet of Things (IoT) devices are being woven into the fabric of our physical world. With this rapidly expanding pervasive deployment of IoT devices, and supporting infrastructure, we are fast approaching the point where the problem of IoT based cyber-security attacks is a serious threat to industrial operations, business activity and social interactions that leverage IoT technologies. The number of threats and successful attacks against connected systems using IoT devices and services are increasing. The Internet of Things has several characteristics that present technological challenges to traditional cyber-security techniques. The Internet of Things requires a novel and dynamic security paradigm. This paper describes the challenges of securing the Internet of Things. A discussion detailing the state-of-the-art of IoT security is presented. A novel approach to security detection using streaming data analytics to classify and detect security threats in their early stages is proposed. Implementation methodologies and results of ongoing work to realise this new IoT cyber-security technique for threat detection are presented.

Huyn, Joojay.  2017.  A Scalable Real-Time Framework for DDoS Traffic Monitoring and Characterization. Proceedings of the Fourth IEEE/ACM International Conference on Big Data Computing, Applications and Technologies. :265–266.

Volumetric DDoS attacks continue to inflict serious damage. Many proposed defenses for mitigating such attacks assume that a monitoring system has already detected the attack. However, many proposed DDoS monitoring systems do not focus on efficiently analyzing high volume network traffic to provide important characterizations of the attack in real-time to downstream traffic filtering systems. We propose a scalable real-time framework for an effective volumetric DDoS monitoring system that leverages modern big data technologies for streaming analytics of high volume network traffic to accurately detect and characterize attacks.