Security

Surveillance and Performance Monitoring of Renewable Plants

Key Research Topics: Thermal-based Human Detection & Tracking, Abnormal Behavior Recognition, Early Defect Detection and Maintenance, Thermal Imaging for Performance Inspection of Photovoltaic Parks

The research group of our Lab is actively involved in developing ICT-enabled emerging technologies for monitoring Plants like Photovoltaic Parks. New algorithms have been developed that utilize the unique characteristics of thermal imaging for both Surveillance/Security and Performance purposes. The algorithms developed employ computer vision and image processing techniques for moving objects spatio-temporal analysis, object classification (e.g. human versus animals) as well as trajectory analysis and suspicious behavior detection for security purposes. The ultimate aim is to minimize false alarms usually encountered in existing security systems increasing their effectiveness and credibility, by fully exploiting the capabilities of thermal imaging. In addition, pioneer research is being performed for analyzing thermal response of PV panels for early diagnosis of potential defects. Algorithms have been developed that allow real-time analysis of individual PV panels in respect of identifying hot spots that are highly correlated with technical defects or other factors (e.g. broken cells or glass, shadows due to the presence of dirt or tree branches, etc). In this context, the ultimate goal is to provide new tools for cost-effectiveness maintenance of such outdoor plants.

PV plant monitoring developed by our lab for real-time operation inspection of the installed equipment

PV plant monitoring developed by our lab for real-time operation inspection of the installed equipment

 

Surveillance Reports are provided by the PV monitoring platform developed by our research team for security evaluation and suspicious behaviour identification

Surveillance Reports are provided by the PV monitoring platform developed by our research team for security evaluation and suspicious behaviour identification.

 

PV monitoring platform developed by our lab, provides performance reports regarding solar panel thermal response utilizing aggregated spatio-temporal thermal information

PV monitoring platform developed by our lab, provides performance reports regarding solar panel thermal response utilizing aggregated spatio-temporal thermal information

Relative Projects: IPv-Park

 

Indoor / Outdoor Security - Surveillance

The research group deals with the development of novel algorithms, technologies and architectures for application in the area of indoor/outdoor surveillance for security reasons. The focus is especially laid on novel approaches for detecting, analyzing predicting (if posssibel) and guiding people and individuals behaviour under several indoor/outdoor environments and a rich variety of contextual conditions. In this respect, state of the art surveillance approaches are developed and applied regarding abnormal and fraudulent behaviour or events, implementing high standards of perimetry security.

Similarly, evacuation planning algorithms are designed and evaluated, supporting among other personalized guidance (e.g. optimal routing even fro disabled people, etc.) via visualization of charts with the corresponding istructions on the mobile devices in dynamically changing conditions (e.g. sudden fires, wall collapsing etc.). 

Last but not least, one of the most recent works of the research group involves the development of technologies for the reduction of the criminal activity via the avoidance of petty crimes in small business environments, based on low cost components and built-in capabilities (i.e. sensors, ad-hoc and embedded systems) interconnecting using established and emerging technologies, such as Digital Subscriber Lines and Cloud computing under a dynamically unified system-umbrella. 

Relevant Projects: SaveMe (FP7), P-React (FP7), IPv-Park (GSRT)

Relevant Publications: 

  1. I. Tsekourakis, C. Orlis, D. Ioannidis and D. Tzovaras, "The Save Me Project Real-Time Disaster Mitigation And Evacuation Management System", in Proc. of 7th IET System Safety Conference incorporating the Cyber Security Conference, 15-18 October 2012, Edinburgh, UK. 
  2. I. Tsekourakis, C. Orlis, D. Ioannidis and D.Tzovaras, “A Decision Support System for Real-time Evacuation Management and Rescue Team Planning during Hazardous Events in Public Infrastructures” , in 12th International Scientific Conference "Transport Systems Telematics", 10-13 October, Katowice-Ustron, Poland, 2012.

 

Cyber Security

As information and communication technologies become more mature, the amount of information generated and exchanged exceeds every previously imaginable limit. Besides the obvious advantages, this fact raises severe security issues that mainly stem from the difficulty in processing big (amounts of) data stemming from heterogeneous sources. 

To this extent, the visual analytics research domain has emerged as a solution for processing, visualizing and reasoning, based on massive amount of data. Analysts are asked to use these tools, so as to rapidly notice the expected and discover the unexpected semantic information within data, whose serial processing time would be orders of magnitude far from real-time.

In this concept the group develops SoA visual analytics techniques, in order to address emerging areas ranging from network information security and attack attribution, to attack detection of BGP (Border Gateway Protocol) hijacking, covering this way, both the tactical (monitoring in real time) and strategic (long term) aspects of security. 

Moreover, the group has recently laid specific focus also on the security issues of the rapidly evolving mobile networks that are becoming increasingly susceptible to attacks.  In this respect, the modus-operandi of the attackers is investigated, analyzed and modelled, by studying both malicious applications and the corresponding traffic to capture traces that reveal the objectives of the attack, its technical details and the attacker. Last but not least, our research focuses on the identification of unauthorized, illicit and suspicious/anomalous behaviour, solely based on network traffic, by monitoring large set of heterogeneous and anonymized data across different network layers (e.g., network transport, service) and possibly network components (e.g. HLR, MSC, etc).

cyber security

Visual Analytics in Network Security

 

cyber security

Signaling/Billing attack Detection techniques in mobile Networks (HLR)

Relevant Projects: VisSENSE (FP7), NEMESYS (FP7)

Relevant Publications: 

  1. E. Biersack, Q. Jacquemart, F. Fischer, J. Fuchs, O. Thonnard, G. Theodoridis, D. Tzovaras, P.-A. Vervier, “Visual analytics for BGP monitoring and prefix hijacking identification”, Special Issue on Computer Network Visualization, IEEE Network Magazine, Nov. 2012
  2. O. Tsigkas, O. Thonnard &D. Tzovaras, “Visual Spam Campaigns Analysis Using Abstract Graphs Representation ", 9th Symposium on Visualization for Cyber Security (VizSec 2012).
  3. O. Tsigkas & D. Tzovaras, “Analysis of Rogue Antivirus Campaigns Using Hidden Structures in k-partite Graphs", 11th International Conference on Cryptology and Network Security (CANS 2012)"
  4. G. Theodoridis, O. Tsigkas and D. Tzovaras, "A Novel Unsupervised Method for Securing BGP against Routing Hijacks", 27th International Symposium on Computer and Information Sciences (ISCIS 2012)"
  5. S. Papadopoulos, K. Moustakas, D. Tzovaras, ”Hierarchical Visualization of BGP Routing Changes Using Entropy Measures”, 8th International Symposium on Visual Computing, July 16- 18, 2012 
  6. S. Papadopoulos, K. Moustakas, D. Tzovaras, ”BGPViewer: Using Graph representations to explore BGP routing changes”, 18th International Conference on Digital Signal Processing (DSP), 1-3 July 2013 
  7. S. Papadopoulos, G. Theodoridis, D. Tzovaras, ”BGPfuse: Using visual feature fusion for the detection and attribution of BGP anomalies”, 10th Symposium on Visualization for Cyber Security (VizSec), October, 2013 
  8. S. Papadopoulos and D. Tzovaras, "Towards Visualizing Mobile Network Data", in Proc. 28th Int. Symp. on Computer and Information Sciences (ISCIS’13), Oct. 2013.

View Related Demos