Fortgeschrittene IT-Sicherheit - WS 2006-07
| 18.483 |
|
Oberseminar: Fortgeschrittene IT-Sicherheit |
|
Organisers: |
Dieter Gollmann (TU), Joachim Posegga |
Time: |
Thursday 18:15-20 (2 SWS)
|
Location: |
Room 221, ESA 1 W - Main Campus
Edmund-Siemers-Allee 1, 20146 Hamburg |
Driving Directions & Room description
Address of lecture hall ESA 1 W, room 221:
Edmund-Siemers-Allee 1
20146 Hamburg
More information on this lecture hall
For driving directions please consult the map on the right or use the more detailed map of Hamburg.
Schedule
| Date |
Speaker |
Location |
Title |
| 26.10.2006 |
Dieter Gollmann, Joachim Posegga |
ESA 1 W,
room 221 |
Introduction |
02.11.2006
|
Mayutan Arumaithurai |
ESA 1 W,
room 221 |
NSIS
implementation on OMNET  |
| 09.11.2006 |
Dimiter Milushev Vladimirov |
ESA 1 W,
room 221 |
Intrusion Detection and Prevention to Enhance Security in VoIP Environments  |
| 16.11.2006 |
Franjo Severinac |
ESA 1 W,
room 221 |
A bridge as a representative exemplified on ISAKMP |
| 23.11.2006 |
Axel Grossklaus |
ESA 1 W,
room 221 |
Policy-Based Management of distributed Network Security Components  |
| 30.11.2006 |
Tilmann Holst |
ESA 1 W,
room 221 |
Automatic Correlation, Rating and Analyzing of
Heterogeneous Network and Incident Data  |
07.12.2006 |
Siddharth Somasundaram |
ESA 1 W,
room 221 |
Web Services Firewalls  |
| 14.12.2006 |
Björn Zessack
Malko Steinorth |
ESA 1 W,
room 221 |
AES on sensor nodes 
Automated Penetration Tests  |
| 21.12.2006 |
Úlfar Erlingsson (Microsoft Research Silicon Valley) |
ESA 1 W,
room 221 |
XFI: Software Guards for System Address Spaces  |
| 28.12.2006 and 04.01.2007 - Christmas
vacation |
| 11.01.2007 |
Björn Engelmann |
ESA 1 W,
room 221 |
Short talk: Dynamic web application analysis for client-side XSS protection  |
| 18.01.2007 |
Anna Otto
Juliya Tkachenko |
ESA 1 W,
room 221 |
Entwicklung von Datenschutzrollen für das Audit Information System im SAP ERP
Probleme und Lösungsansätze
zum Identity Management System in einer
universitären Umgebung |
| 25.01.2007 |
Iam Sathish
Zheng Lei |
ESA 1 W,
room 221 |
Analysis of Security Features in IPv6
Integrate GPS Tracking Devices into Application with XML  |
| 01.02.2007 |
Saverio Niccolini (NEC Europe) |
ESA 1 W,
room 221 |
SPam over Internet Telephony (SPIT) prevention: state of the art, research challenges and solutions |
| 08.02.2007 |
Björn Ahne
|
ESA 1 W,
room 221 |
Short talk: Dynamic Evaluation of Input Filter Functions |
Contact
Please contact Henrich C. Pöhls for further questions and for scheduling your talk.
Abstracts
02.11.2006 - Mayutan Arumaithurai - NSIS
implementation on OMNET (Slides )
Recently on the internet there is an increase in the number of real-time
services which are time-sensitive and require consistent performance with
respect to loss, delay and jitter. Best-effort delivery is not adequate to
support it. Therefore to achieve predictable performance levels, a QoS
signalling technique is required to reserve and manage network resources.
The proposed QoS signalling protocol RSVP has not been able to scale well
and therefore after a lot of efforts to improve RSVP the focus was shifted
to NSIS (Next Steps in Signalling). As NSIS is a relatively new protocol,
there is a lot of scope for improvement towards a successful deployment. For
this purpose NSIS has to be tested on a large scale.
09.11.2006 - Dimiter Milushev Vladimirov - Intrusion Detection and Prevention to Enhance Security in VoIP Environments (Slides )
The initial hype about VoIP technology deployment, resulting mainly from the promise of great cost savings, has gradually turned into hype about VoIP securrity issues. T he solution of the se security issues is expected to be critical for further large scale deployment and development of the technology. In this talk, I explore the possibility of enhancing and utili sing an existent and deployed network security technology, namely Intrusion Detection and Prevention , in order to considerably improve the security level in a VoIP environment . I will present a SIP IDS prototype software, developed by NEC labs, Europe. Then I discuss my work, which mainly consists of evaluating and enhancing the prototype in order to counter additional VoIP-related attacks, such as REFER, re-INVITE, REGISTER-related attacks etc. The conclusion of my talk is that the theoretical foundation for the usage of IDS for VoIP is relatively sound , but very general and therefore specific implementations should be evaluated to determine how feasible the IDS approach for VoIP environments is .
16.11.2006 - Franjo Severinac - A bridge as a representative exemplified on ISAKMP (Slides )
Network traffic between any (two) interconnected hosts can be secured by the use of IPsec. But it is not always possible or
economical to upgrade the software and configuration of all nodes participating in the network. One solution to this problem is
the use of a bridge as a security gateway, which applies the relevant security protocols on behalf of the protected nodes.
The OpenBSD Ethernet bridge is extended to adopt a hosts identity on ISO OSI network layer and data link layer, with
the aim, to be totally transparent to a protected host and the destination host or gateway. ISAKMP uses this functionality
to exchange key material and afterwards to encrypt totally transparent. The presentation introduces the concept of the
bridge being a representative.
23.11.2006 - Axel Grossklaus - Policy-Based Management of distributed Network Security Components (Slides )
Recent changes in the area of network use and network security demand moving
from a central approach to a distributed one when deploying network security
components in a network. This leads to an increase in costs and complexity
of management and demands more methodology on how to get the best possible
security benefit out of the invested resources.
At the same time, policy based network management not only offers a powerful
framework for central network management but outright demands control over
each and every aspect of every system in a network.
How can these two techniques be combined? What primary and secondary goals
should be observed and how can an actual, and possibly even measurable,
security gain be achieved?
The talk describes a first step towards the development of a
formally-founded management framework that is trustworthy and at the same
time able to deal with the requirements, imperfections and general
conditions of large-scale real-world networks.
30.11.2006 - Tilmann Holst - Automatic Correlation, Rating and Analyzing of
Heterogeneous Network and Incident Data
(Slides )
In recent years an enormous increase in IT security incidents has been
noticed. The amount of directed serious attacks is almost constant, but
the amount of automated unselective attacks is growing exponentially.
This thesis will use a top-down approach to develop a method to
effectively handle security incidents caused by mass attacks. It will
start by introducing a common language for such incidents. Based on that
common language different types of incidents will be described and
categorized. As a last step requirements to detect these different
incident types will be given and tested against information sources
currently available - or needed.
7.12.2006 - Siddharth Somasundaram - Web Services Firewall
(Slides )
Motivation: Network firewalls cannot detect web services threats because the ports that the web service operates, are open to HTTP traffic thereby allowing SOAP messages to flow undetected into a network. An attacker can put malicious command or data into the web service. As a normal firewall provides no message security at all, other ways of securing messages are necessary.
Solution: Web Services Firewall is an XML Application Firewall that works at the application level with in depth knowledge of web services. Web Services Firewall checks the content of the SOAP messages and thereby deciding whether the SOAP message should be delivered to the target operation of the target web service. Web services firewall cannot create secure web services but can only secure existing web services by managing the threats.
14.12.2006 - Björn Zessack - AES on sensor nodes (Slides )
The key feature of sensor nodes are their small size, their low power consumption and their ability to communicate wirelessly. Therefore, sensor nodes can be used in various application domains. This talk focuses on the monitoring of ship containers. Data monitored by the sensor nodes is sent via a base station to a data warehouse. Since the transmitting power of such sensor nodes is limited, data has to be relayed to reach the base station. For privacy reasons, the relaying nodes must not be able to understand data from other containers. For this reason, the data must be encrypted before sending.
Encryption on sensor nodes is challenging due to memory and power constraints. However, the AES algorithm can be used because of its adaptability. It was implemented on a Scatterweb ESB 430 sensor node which is distributed by the FU Berlin. The talk describes the experiences of implementing AES on the ESB 430. Also, various adjustments are discussed.
14.12.2006 - Malko Steinorth - Automated Penetration Tests (Slides )
Penetration testing is a well known method to measure the security
level of a network. Nevertheless there are disadvantages like high
prices and the fact that only a snapshot of the security level is
shown.
By identifying sub processes of a penetration test that can be automated these drawbacks can be addressed. Due to automation penetration tests can be conducted at a more favourable price and therefore more often. Of course valuable information might get lost by automating penetration tests. Thus automated penetration tests can only be an add-on for other methods evaluating the IT security.
Based on existing penetration testing methodologies this thesis will work out penetration test tasks that can be automated. On this basis a way to aggregate and report this data will be developed. Special focus will be on the reporting of the ongoing development of the security level over time.
As this thesis follows a practical approach a prototype will be implemented and applied to example networks to assess the developed techniques.
21.12.2007 - Úlfar Erlingsson - XFI: Software Guards for System Address Spaces (Slides )
XFI is a comprehensive protection system that offers both flexible access
control and fundamental integrity guarantees, at any privilege level and
even for legacy code in commodity systems. For this purpose, XFI combines
static analysis with inline software guards and a two-stack execution model.
We have implemented XFI for Windows on the x86 architecture using binary
rewriting and a simple, stand-alone verifier; the implementation's
correctness depends on the verifier, but not on the rewriter. Our
experiments confirm that XFI offers pervasive protection with only modest
enforcement overheads. We have applied XFI to software such as device
drivers and multimedia codecs; the resulting modules function safely within
both kernel and user-mode address spaces. This is joint work with Martin
Abadi, Michael Vrable, Mihai Budiu, and George Necula.
Speaker's BIO:
Úlfar Erlingsson is a researcher at Microsoft Research's Silicon Valley
Center. He did his graduate work in the mid-to-late 90's at Cornell
University's Information Assurance Institute on specifying and enforcing
security properties using program rewriting techniques. Later, he was
director of privacy protection for deCODE Genetics, where he oversaw the
security design of a centralized healthcare database. After this, he
co-founded and was CTO of Green Border Technologies, a Silicon Valley
security software company. Recently, Úlfar's research has focused on
low-level security mitigation and dependability techniques (e.g., involving
hypervisors, hardware devices, and the precise syntax and semantics of x86
opcodes) in the Gleipnir project
(http://research.microsoft.com/research/sv/gleipnir/).
11.01.2007 - Björn Engelmann - Dynamic web application analysis for client-side XSS protection
(Slides )
One of the main reasons for the recently observed rapid increase of
disclosed XSS vulnerabilities is the classical "misplaced incentives" situation: While the problem is caused by defective web applications, the ones jeopardised are their users. This fact, along with coverage considerations clearly demands a client-side defense mechanism, be it only as a last line of defense. This talk will describe the initial steps towards a formal framework for script-focussed structural website analysis able to distinguish an XSS attack from a change in the application. Primary objective is a tool using data-mining techniques to derive security-relevant information from observed user interactions.
18.01.2007 - Anna Otto - Entwicklung von Datenschutzrollen für das Audit Information System im SAP ERP
Unter dem Stichwort "Compliance" werden im Wirtschaftsrecht diejenigen Ge- und
Verbotsvorschriften zusammengefasst, die bei der täglichen Arbeit zu beachten sind
und eine entsprechende Kenntnis sowie organisatorische Voraussetzungen erfordern.
Nachdem in den letzten Jahren einige Konzerne in Europa und und den USA wegen
Bilanzmanipulation in die Schlagzeilen geraten sind, sollen die stetig steigenden
Anforderungen der Gesetzgeber die Sicherheit und das Risiko eines Unternehmens
optimieren. Bei Nichtbeachtung dieser Regelungen, können betroffene Mitarbeiter
bzw. die Unternehmensleitung mit hohen Bußgeldern oder sogar strafrechtlichen
Sanktionen belangt werden. Zunehmend von Bedeutung für die "Compliance" sind
auch die Informationssicherheit und der Datenschutz. Die Auditierung von
IT-Systemen ist ein Werkzeug um die Erfüllung der Regulierungsanforderungen
sicherzustellen. In diesem Zusammenhang hat die SAP AG unter anderem das
Audit Information System entwickelt. Dieser Vortrag erläutert die Entwicklung von
Datenschutzrollen für das Audit Information System im SAP ERP.
18.01.2007 - Juliya Tkachenko - Probleme und Lösungsansätze zum Identity Management System in einer universitären Umgebung
In den letzten Jahren wurde eine große Anzahl verschiedener neuer Anwendungen und Systeme in das IT-Umfeld vieler großer Organisationen, Unternehmen und Behörden eingeführt. Dementsprechend stieg der Aufwand, der von solchen Organisationen für die Verwaltung der Benutzerdaten auf verschiedenen Systemen gefordert wird, immer weiter an. Jedes System in einer so heterogenen Landschaft hat meistens eine eigene Benutzerverwaltung und eigene Administratoren, sodass hier oft neue „Silos“ von Identitätsinformationen erzeugt werden. Somit kann die Komplexität der Benutzerdatenverwaltung, die proportional zu der steigenden Anzahl der Identitätsinformationen in einer Organisation immer größer wird, irgendwann zum Verlust der Kontrolle über Organisationsdaten und -prozesse führen.
Ein Identity Management System soll diese Probleme lösen, indem die Verwaltungsaufgaben für digitale Identitäten über alle Systeme einer Organisation integriert und automatisiert werden. Dabei werden verschiedene Daten miteinander verknüpft, welche zu einer Person in verschiedenen Systemen einer Organisation gespeichert sind. So entstehen einheitliche digitale Identitäten, die man zentral verwalten kann. Dadurch lassen sich Kosteneinsparungen realisieren, Bearbeitungszeiten reduzieren und Fehlerquellen minimieren.
25.01.2007 -
Iam Sathish - Analysis of Security Features in IPv6
This project aims at studying and analyzing the security features in Internet Protocol version 6 (IPv6). The various limitations in IPv4 has enabled the researchers to migrate from IPv4 to IPv6. With the growing popularity in the Internet and number of users increasing day by day, IPv6 has gained significant importance with various new features and enhancements which weren't found in IPv4. There are various new capabilities implemented in IPv6 such as Addressing and routing enhancements, simplification of Header format, new QOS capabilities and more importantly Authentication and Privacy capabilities.
The GOOD news is that security in IPv6 is very much like security in IPv4 but the BAD news is that security in IPv6 is very much like security in IPv4. This project deals with the study of why we go for the new version of the protocol IPv6, what are the flaws in IPv6, security framework implemented in IPv6 analysing in detail the various header formats and structures, the various IPv6 core protocols and their security considerations, information about what are the issues involved in the implementation of Firewall in IPv6, a detailed description of various IPsec Policy Modeling concepts, how the researchers have taken care of the transitions from version four to version six, the various threats and defences in IPv6 and steps taken care for the development of IPv6.
25.01.2007 -
Zheng Lei
- Integrate GPS Tracking Devices into Application with XML (Slides )
GPS Tracking devices are used to track, monitor and report on time and vehicle/person activity, including real time information on operational status (i.e. loading / unloading, in transit, waiting, return to base, current environment temperature, person motional status etc).
Devices talk with Servers with plaintext communication protocol. The project task is to use XML technology to build a dynamic GUI generation framework to integrate GPS tracking devices of different wireless communication technology and communication protocol.
01.02.2007 -
Saverio Niccolini
-
SPam over Internet Telephony (SPIT) prevention: state of the art, research challenges and solutions
Spam is defined as the transmission of unsolicited mails; it is considered to be one of the biggest problems the Internet has ever faced. Today, more spam emails than regular emails are transmitted in the public Internet. Among other reasons, the spam problem became so widespread because there were no solutions developed ubiquitously before the problem arose. Nowadays there are available methods able to counter this problem with different approaches, but none of these methods constitutes a definitive solution.
With the increasing deployment of Internet telephony solutions, it is commonly expected that a similar form of spam will show up in this area. This threat is commonly referred to as SPIT (SPam over Internet Telephony) or Voice over IP (VoIP) spam. SPIT is defined as the transmission of unsolicited call over Internet telephony.
This talk analyzes the requirements for SPIT prevention, provides a thorough classification of currently known SPIT prevention methods, and introduces a generic SPIT prevention system architecture. As an instance of the generic architecture we designed and implemented an advanced SPIT prevention system composed of methods that avoid unnecessary callee interaction and fulfill the requirements of being less intrusive as possible from an end-user point of view while remaining adaptive in order to be customized for different scenarios.
Speaker's BIO:
Dr. Saverio Niccolini is currently a Senior Research Staff Member at NEC Europe Ltd., Network Laboratories in Heidelberg, Germany. He received his Ph.D. in Information Engineering in 2004 from University of Pisa with a thesis on "Voice over IP: Traffic Control, Resource Allocation and Measurements Techniques in Next Generation Networks". He authored two books (“IP Telephony Cookbook” and “Guide to Network Resource Tools”) and about 20 papers published in international conferences and magazines. His research interests include Voice over IP, Next Generation Networks architectures, security management and measurement techniques. He has been TPC co-chair of the VoIP Management and Security workshop collocated with IEEE NOMS in Vancouver in April 2006 and currently active in a number of other TPCs.
08.02.2007 - Björn Ahne - Dynamic Evaluation of Input Filter Functions
Web-applications have become exposed to a constantly growing palette of attack options. Cross Site Scripting, SQL Injection, Path Traversal and many more. Many of this attacks are only possible if user inputs aren't filtered carefully enough. But writing filters that check inputs for the whole bandwidth of attacks in all their variations isn't an easy task as it may seem at first glance. Exceptions to restrictions, failure tolerance of browsers and different kind of encodings make it really a complex task with many chances of failure.
This short talk presents a little introduction into a diploma thesis which aim it is, to help developers write proper input filters through an automated evaluation process. The talk will start by describing the problem motivating this work and continues with announcing the goals and presenting the structure (of the software) used to archive them.
______________________________
Auszug aus dem KVV:
Sicherheitsrelevante Themen von Informatiksystemen werden anhand von aktuellen Projekt-, Baccalaureats- und Diplomarbeiten und Dissertationsvorhaben vorgestellt und intensiv diskutiert. Das Seminar wird gemeinsam vom Arbeitsbereich SVA der Technischen Universität Hamburg-Harburg und dem Arbeitsbereich SVS des Fachbereichs Informatik der Universität Hamburg angeboten und soll auch dem Austausch der aktuellen Forschungsergebnisse zwischen den beiden Arbeitsbereichen dienen und die Zusammenarbeit vertiefen.
Das Oberseminar soll Studierenden und Promovierenden, die sich im Fachgebiet IT- Sicherheit vertiefen, eine Plattform bieten, um aktuelle Probleme und Forschungsarbeiten im Bereich IT-Sicherheit vorzustellen und zu diskutieren. Gastvorträge sind im Programm ebenfalls vorgesehen.
|