CATEGORIES:
BiologyChemistryConstructionCultureEcologyEconomyElectronicsFinanceGeographyHistoryInformaticsLawMathematicsMechanicsMedicineOtherPedagogyPhilosophyPhysicsPolicyPsychologySociologySportTourism
CrISS 2013 TIMETABLE
Dr Iosif Androulidakis, University of Ioannina, Greece, “When VIRII use phone catalogs to uncover telecom critical infrastructure” Private telephony networks are used by Hospitals, Ministries, Banks, Public and Private bodies/authorities, Companies, Industry, Police, Army and practically most of a nation’s critical infrastructure entities. Thus they effectively are themselves part of a nation’s critical infrastructure. In previous DESSERT workshops, we have highlighted their importance and have proposed a targeted, centralized project comprising of educational, policy, auditing, technical, documentation, hardware and software solutions and actions in order to both secure them and educate their users. This year, we will briefly describe the idea of a virus targeting private telephony networks. We will also highlight the fact that public phone catalogs can provide information valuable to attackers, effectively uncovering telecom critical infrastructure.
Dr Eugene Brezhniev, National Aerospace University KhAI, Ukraine, “Multi-factor analysis of critical infrastrusuture safety: BBN-based approach” Key challenges related to critical infrastructure (including NPP-based power grid) safety assessment are analysed. Principles and technique of Bayesian Belief Network (BBN)-based analyse of critical infrastructure safety taking into account complexity, uncertainty are described. Application of the technique is illustrated by use of Fukushima accident analysis. Prof Dr-Ing Thomas Hollstein, Tallinn University of Technology, Estonia, “Dependable many-core architectures for mixed-criticality applications” Currently computer architectures are undergoing one of the largest paradigm shifts since the invention of the microprocessor. The scaling of classical computer architectures (based on concentrated multi-core processors) is limited due to the bottleneck of sharing one common DRAM interface. New developments in IC manufacturing technologies allow 3D chip stacking, using Through-Silicon-Vias (TSVs). This enables distributed shared memory (DSM) architectures for NoC based scalable many-core architectures, providing local DRAM access for every processor tile. Scalability, sophisticated dependability concepts and suitable programming models are essential requirements for successful introduction and application of these new architectures. In this presentation we will approach the whole topic area from the point of view of the dependability service layer and application mapping under mixed criticality constraints. From this perspective we will draw conclusions on necessary implications related to NoC properties, dependability concepts and application deployment on many-core architectures.
Dr Artur Jutman, Tallinn University of Technology, Estonia, “Embedded instrumentation infrastructure for ageing failure resilience” Semiconductor products manufactured with latest and emerging processes are increasingly prone to wear out. Ageing in integrated circuits has already become a real issue and is considered to be of vital importance in the future – especially for the increasing number of safety-critical applications in the automotive, avionics and medical arena. There are few options to reduce ageing at the technology level. Hence, semiconductor technology leaders are seeking for efficient ways to prolong the products lifetime using special system and circuit design techniques improving their fault tolerance and failure resilience characteristics. Fault management becomes an important function performed by the system that is aimed at graceful degradation. The talk, in the first place, analyses requirements for an efficient fault management system and makes a conclusion regarding the way the architecture of the service infrastructure can be implemented. The architecture we propose is based on the latest upcoming DFT standard IEEE 1687 IJTAG, but we define a set of additional design rules that have to be implemented to facilitate automatic integration of instrumentation networks from third party IP cores into a single SoC-wide homogenous infrastructure, characterized by a very small error detection latency (practically immediate fault detection) and logarithmic time to localize the faulty resource, which is confirmed by experimental results.
Dr Vitaly Levashenko, University of Zilina, Slovak Republic, “Data mining support for analysis of critical infrastructures” Data mining is a process of extracting implicit, potential, novel, useful and intelligible patterns from mass data of data sets, databases or data warehouse, etc. The technologies of classification, estimation, prediction, affinity grouping, association rules, clustering, description and visualization are covered in data mining, which is widely used in the different fields. Decisions play an important role in analysis of critical infrastructure too. A decision tree is a graphic model of a decision process, and it is usually used as a decision support tool or classifier. A decision trees is one of the best ways to analyze a decision, as it is visualized and simple to understand and interpret.
DrSc, Prof Aleksander V. Potii, Kharkiv National University of Radio Electronics, Kharkiv Air Force University, Ukraine, “Lightweight (green) cryptography: modern state and perspective of development” À key goal in the development of security mechanism for extremely low cost platforms, such as passive RFID tags, SCADA device, machine-to machine communication in Internet of things, implantable medical devices is the design of low cost security primitives that consume very little power (in the order of few microWatts) and have adequate performance (throughputs of hundreds of kilo-bits per second).After several attempts to integrate traditional cryptographic primitives into small, embedded, and extremely resource constrained devices, the results were mostly unsatisfactory. As a result, a new branch of cryptography, commonly called Lightweight Cryptography (or Green Cryptography), emerged to address the issues of these tiny ubiquitous devices. In our presentation we discuss the need for Lightweight Cryptography and modern state in this area of research. Prof Dr Harald Richter, Clausthal University, Germany, “On a reliable real-time computer network for safety-critical applications” In this talk, a reliable real-time computer network for safety-critical applications is presented. The network comprises all 7 layers of the ISO-Modell and contains redundancy and protocol measures on these layers to achieve high reliability under real-time constraints. The network operates at 3.125 Gb/s, its latency between neighbor nodes is in thems-range. It consists of rings as topological elements which are coupled by routers. The intended application is for in-car communication in order to replace bus systems such as CAN and Flexray. All functions of a car such as steering, breaking or clutching can be transmitted from electronic controller units to actuators via data packets. Other application areas are in process and factory automation. Dr Elena Zaitseva, University of Zilina, Slovak Republic, “Reliability analysis of multi-state system” There are two mathematical models for the interpretation of an initial system in reliability engineering that are named a Binary-State System (BSS) and a Multi-State System (MSS). The BSS is used for mathematical representation of system with two states: failure/unavailable (indicated as 0) and functioning/available (indicated as 1). The MSS allows defining more than two states (performance levels) of the system: from complete failure (represented by 0) to perfect functioning (indicated as M –1), where M is a number of the system performance levels. Every MSS component can take some states too. Some methods for the MSS estimation based on Multiple-Valued Logic will be considered. Date: 2015-12-18; view: 693
|
KEYNOTE SPEAKERS