Publications

A list of my publications.

 

1.     The mTeam project

The development of powerful, mobile devices, equipped with Internet connectivity and fast wireless local area networks, creates new opportunities for collaborative work. In the modern Knowledge Economy, mobile, nomadic knowledge workers work in flexible teams on projects that require a creative approach. The goal of the mTeam project is enabling the creation of applications that would exploit the full potential of mobile devices in the support of collaborative work. The realization of this goal requires the fulfilment of several requirements:

• Since the users of mobile devices do not always have Internet connectivity, collaborative applications should be distributed and able to work even in the absence of a fixed infrastructure.
• Security, privacy and trust management are a crucial requirement for users of collaborative applications.
• True collaboration exists only in a social context that should be expressed and used by the supporting applications.

The mTeam project will focus on the algorithmic foundations and basic research problems in realizing such decentralized collaborative applications, as well as systems issues like creation of middleware that can be used to support these applications. For the purpose of middleware evaluation, two applications will be developed that will exploit the entire spectrum of middleware functions: the TeamCreator and the TeamEnvironment.

For more information, see: mTeam.pjwstk.edu.pl

2.     The uTrust project

The goal of the uTrust (Universal Trust) project is to develop a universal and formalized approach for trust management in a wide range of distributed information systems. Basing on this approach, the practical goal of the project is to provide a universal library of trust management functions. In the future, trust management should become yet another, standard service of information security, such as authentication, authorization, privacy or integrity. For this to happen, it is necessary to define standard, universal primitives of trust management. Existing approaches for managing trust are often tied to particular domains, too complex to compute or based on arbitrary assumptions. The uTrust project will try to overcome this limitation by using a set of standard benchmarks and adversary models to test universal algorithms and protocols. The uTrust library will be developed as an open-source project.

For more information, see: uTrust.pjwstk.edu.pl

 

3.     Social Informatics

I define Social Informatics as an area of informatics that attempts to realize social goals (such as fairness), apply social concepts (such as trust), or become a source of information for social analysis. In informatics and computer science today, there is an increased interest in such a research area. Examples include Webmining, computational social choice, social simulation, Web science as defined by Tim Berners-Lee, trust management, computational social networks and many other areas. My own research in that area focuses on trust and fairness in open, distributed systems.

 

4.     Peer-to-peer games

We are working on mechanisms to provide fairness in Peer-to-Peer (P2P) games without using centralized coordination. For example, we have developed a protocol for fair drawing from a finite set (with applications in various games of turns, such as bridge, poker, Scrabble etc.). The protocol does not use centralized resources, but makes it impossible for players to cheat. Currently, we are researching the subject of scalable and fair design of massive, multiplayer P2P games.

 

5.     Reputation mechanisms for e-commerce

Reputation mechanisms are used for trust management in many applications of e-commerce, such as auctions or Internet stores. We are interested in evaluating the performance of these mechanisms in environments where users can collaborate (form coalitions) for unfair use of the application (for instance, fraud in Internet auctions). These coalitions are used to manipulate the reputations of users. In systems without central coordination, such as P2P systems, the design of reliable reputation systems is especially difficult. In contrast to previous research, we try to model the reputation and e-commerce system as realistically as possible, without using simplifying or paradigmatic assumptions from game theory.

 

6.     Security of wireless 802.11b,g networks

The widely used protocol for information security in WiFi (802.11) networks, Wireless Equivalent Privacy, has cryptographic weaknesses that can be exploited to break security. One of the possibilities for increasing its security is a more frequent exchange of keys (proposed in the WPA standard, however not yet widely used). We have developed a method and an advanced prototype that allows for automatic, periodic and secure exchange of keys for WEP. Currently, we are working on a secure access point for 802.11g, designed on available hardware and free software with an emphasis on operating system and application security.

 

7.     Ad-hoc networks based on the Bluetooth standard

The Bluetooth standard envisions, apart from personal area networks, the creation of ad-hoc networks that can have a larger range and connect more devices. The standard does not, however, specify the protocols of routing and self-organization in such networks. We are working on ad-hoc routing protocols for Bluetooth networks that attempt to exploit existing fixed network infrastructure.

 

8.     A standard and translator for signatures of Intrusion Detection Systems

We have proposed a standard for signatures of Intrusion Detection Systems (IDS) (the standard, called CIDSS, is described in an Internet Draft). The proposed Common Intrusion Detection Signature System (CIDSS) is based on experiences of automatic translation of signatures from various IDS vendors. The standard can be thought of as a “sum” of the expressive capabilities about intrusion attempts that are available in various IDS systems. CIDSS is used as a basis for the development of tools for automatic IDS signature translation. Currently, signatures from 4 systems: Snort, Shoki, RealSecure and Dragon are translated into and from CIDSS. Among many potential applications, we can mention the creation of signature databases and of audits of IDS system applications.

 

9.     Methods for e-learning of movement

We have been the first to propose the use of advanced e-learning methods for teaching sport disciplines such as dance, martial arts or rehabilitation (more generally, for teaching of movement). The use of such methods requires solving many problems, such as automatic evaluation of the execution of movement by a student, adaptation of the teaching material to a student profile, and many others. We are researching solutions for these problems, based on a case study that aims to create e-learning material for the teaching of a short form of the Chinese martial art of wu-shu. In the case study, we are cooperating with the trainer of the Polish national team of this discipline.

 

10.     Tools for the support of knowledge creation

The research of knowledge creation, both by an individual and in organizations, is an important new topic (particularly in Japan). We are focusing on the design of supporting tools for knowledge creation in universities or research laboratories. The tools will be collaborative, allowing the organization of debates or brainstorms, and the support of knowledge creation will be based on contemporary theoretical developments in the area.