Adaptive Computing Research Group
The group conducts research on modelling and optimization in resource-constrained distributed systems. Our work concentrates on data analysis, probabilistic modelling and component-based development for mobile context-aware systems, and algorithmic aspects of ad hoc and sensor networks, especially local algorithms. We work in a multidisciplinary environment in close collaboration with other research groups of HIIT and in national and international collaborative projects.
Our work on context-aware systems is part of the ubiquitous computing vision, in which computational resources are everywhere unnoticeably available. Here "context" refers to the surrounding situation, automatically to be used in applications. Our work in this area covers user modelling, automatic identification of places meaningful to users, tools for gathering sensor data and adaptive user interfaces.
Trust and reputation development is also studied in the group, e.g. using game theory.
Concerning ad hoc and sensor networks, we have studied routing, data gathering, sensor network design and scheduling of actions of the nodes in the network. The work in 2007 has focused on local algorithms. These have many other application areas than sensor networks. A local algorithm is a distributed algorithm where each node operates solely based on the information that was available at system startup within a local neighbourhood of the node.
Contact person: Patrik Floréen
Homepage: http://www.hiit.fi/adaptive-computing
Projects
Personalised Ubiservices in Public Spaces (PUPS)
atankeruun optimointi resurssirajoitteisissa verkoissa (Geru)
Widgets Sharing (WiSh)
Semantic Interpreter Widened Experience (Stepwise)
Trust For All (Trust4All)
Selected publications
J. Suomela: Approximability of identifying codes and locating-dominating codes. Information Processing Letters 103 (2007), 28-33.
P. Nurmi: Perseus: a personalized reputation system. Proc. IEEE/WIC/ACM International Conference on Web Intelligence (WI, Fremont , California , November 2007).
P. Floréen, P. Kaski, T. Musto, J. Suomela: Local approximationalgorithms for scheduling problems in sensor networks.? Proc. 3rd International Workshop on Algorithmic Aspects of Wireless Sensor Networks (Algosensors, Wrocaw , Poland , July 2007). Lecture Notes in Computer Science 4837. Springer-Verlag , Berlin , 2008, 99-113.
P. Nurmi, J. Kukkonen, E. Lagerspetz, J. Suomela, P. Floréen: BeTelGeuse: a tool for Bluetooth data gathering. Proc. 2nd International Conference on Body Area Networks (BodyNets, Florence , Italy , June 2007).
P. Nurmi: Reinforcement learning for routing in ad hoc networks. Proc. 5th International symposium on Modeling and Optimization in Mobile , Ad Hoc, and Wireless Networks (WiOpt, Limassol , Cyprus , April 2007).