The revolutionary development of the Internet of Things has triggered a huge demand for Internet of Things devices. They are extensively applied to various fields of social activities, and concerning manufacturing, they are a key enabling concept for the Industry 4.0 ecosystem. Industrial Internet of Things (IIoT) devices share common vulnerabilities with standard IoT devices, which are increasingly exposed to the attackers. As such, connected industrial devices may become sources of cyber, as well as physical, threats for people and assets in industrial environments. In this work, we examine the attack surfaces of a networked embedded system, composed of devices representative of those typically used in the IIoT field. We carry on an analysis of the current state of the security of IIoT technologies. The analysis guides the identification of a set of attack vectors for the examined networked embedded system. We set up the corresponding concrete attack scenarios to gain control of the system actuators and perform some hazardous operations. In particular, we propose a couple of variations of Mirai attack specifically tailored for attacking industrial environments. Finally, we discuss some possible

No abstract

No abstract

No abstract

Machine Learning (ML) has seen massive progress in the last decade and as a result, there is a pressing need for validating ML-based systems. To this end, we propose, design and evaluate CALLISTO-a novel test generation and data quality assessment framework. To the best of our knowledge, CALLISTO is the first blackbox framework to leverage the uncertainty in the prediction and systematically generate new test cases for ML classifiers. Our evaluation of CALLISTO on four real world data sets reveals thousands of errors. We also show that leveraging the uncertainty in prediction can increase the number of erroneous test cases up to a factor of 20, as compared to when no such knowledge is used for testing.CALLISTO has the capability to detect low quality data in the datasets that may contain mislabelled data. We conduct and present an extensive user study to validate the results of CALLISTO on identifying low quality data from four state-ofthe-art real world datasets.