|
Corvara (Bolzano), Italy
January, 19th - 2010
WORKSHOP ORGANIZERSMassimo Mecella, Sapienza Università di Roma, ItalyCarola Aiello, Sapienza Università di Roma, Italy
PROGRAM16.15 Massimo Mecella - Opening
16.30 Andrea Massa - Real-Time WSN-Based Localization of Passive Targets in a Domotic Scenario
In the recent years, there has been a wide and rapid diffusion of wireless sensor networks (WSN), due to their reliability, increased lifetime and flexibility, and low deployment costs [1]. Due to their wide spreading, WSNs ave been profitably used in several realtime domotic applications, including location and tracking problems. In such a framework, the main efforts have been devoted to develop ad-hoc systems based on dedicated transponders/sensors on the target. Although easier than a "passive" localization technique, the main drawback of these approaches is the need of the target to be equipped with an ad-hoc device. Such a limitation is of particular importance when dealing with people (e.g., privacy) and especially with non-cooperative subjects as for elderly people. Moreover, security applications in a domotic scenario (e.g. access monitoring) cannot be based on active localization technologies.
Other strategies concerned with transceiver-free targets have been also presented in the scientific literature. State-of-the-art approaches are based on Doppler radar systems able to estimate the distance between the target and the sensor. Unfortunately, the arising performance in real environments can be strongly influenced by non-negligible instabilities leading to several false alarms. Furthermore, slow-moving targets are not generally detected.
This contribution is aimed at presenting a new methodology for the localization and tracking of passive objects starting from the measurements of the RSS indexes available at the nodes of a WSN. Such an approach is motivated by the fact that any target lying within the environment interacts with the electromagnetic waves radiated by the nodes. Therefore, the measurements of the perturbation effects on the other receiving nodes are dealt with a suitable inversion strategy. By virtue of the fact that the number of nodes in a WSN can vary and the need to have a simple and flexible tracking/localization method allowing real-time estimates, a learning-by-examples (LBE) strategy is used. Simulated as well as experimental results are provided in a domotic scenario in order to evaluate the reliability of the proposed approach.
17.15 Chun Yu Chen - The Monitoring and Simulation of Home Energy
In spite of the standby power, power usage of a home is directly linked with appliance operation as well as the activities of residence. In order to save energy from wasting, it is necessary to understand the power demand of appliances and when for further reasoning. To achieve this aim, ITRI has developed its own technologies, such as meter plug, home energy controller and energy aware appliances, for power demand monitoring, and the integration of these technologies is the so called home energy management system (HEMS).
To further improve the performance systematically, a machine learning function for demand simulation is also added to this system. This function can help the HEMS to integrate with Advanced Metering Infrastructure (AMI) latterly for better energy efficiency. This function can also help to identify the activities of residence.
18.00 Maurizio Lenzerini - Ontology-based data integration
We present a comprehensive approach to, and a complete system for, ontology-based data integration. In this system, the global schema is expressed in terms of a TBox of the tractable Description Logic DL-Lite, the sources are relations, and the mapping language allows for expressing GAV sound mappings between the sources and the global schema. The mapping language has specific mechanisms for addressing the so-called impedance mismatch problem, arising from the fact that, while the data sources store values, the instances of concepts in the ontology are objects. By virtue of the careful design of the various languages, our system is able to answer unions of conjunctive queries through a very efficient technique (LOGSPACE with respect to data complexity) which reduces this task to standard SQL query evaluation. We also show that even very slight extensions of the expressive abilities of the system lead beyond this complexity bound.
18.45 Domenico Lembo - Efficient query answering over ontologies: The DL-Lite family
In this talk we provide a gentle introduction to the notion of ontology and to Description Logics (DLs) as a formal language
to specify ontologies. We concentrate on those cases in which ontologies are used to access data, and on query answering, which is a crucial task in ontology-based data access. In particular, we present DL-Lite, a family of DLs which provide efficient query access to large amount of data (possibly stored in a RDBMS), and at the same time allow for expressing ontology constructs typical of data modeling formalisms such as the ER model and UML class diagram. We discuss query answering and other forms of reasoning in DL-Lite.
Click here to download the program / Click here to download the presentations. |
Activities 