Adaptive procedures for the assessment and improvement of metrological characterics of sensor networks

COFIN 2004 Petri DIT-PRJ-05-028

Status NOT active project
DISI role Coordinator
Project type Research Project
Dimension National
Acquisition date 2004-11-17
Start date 2004-12-01
End date 2006-11-30

Project details

Project astract The proposed research program is aimed at studying, designing and validating a set of methodologies and/or innovative processes tovmanage the metrological confirmation of sensor networks as well as to assess the decisional risks based on measured data.<br/>In recent years, sensor networks have become one of the main research topics in Information Engineering. Nevertheless, up to now most research efforts have been focused on solving the technological problems concerning with the communication between the network nodes, while partially disregarding the actual quality of the information collected by the sensors. Such problems are instead<br/>of outmost importance whenever acquired data are used in decision-making processes. In fact, managing efficiently and effectively the whole measurement system is essential. This result can be achieved only if suitable strategies for metrological confirmation are applied.<br/>To this purpose, the proposed research program intends to provide valid solutions to the following metrological issues:<br/>- optimal management of the metrological confirmation status of measurement systems based on sensor networks;<br/>- evaluation of the measurement uncertainty effects on decision support procedures and definition of optimal decisional rules.
Keywords Metrological confirmation, Sensors, Learning Systems, Programmable Logic Devices
Fundings 156800 €
Partners
  • DIT - UniTN
  • Polytechnic of Milan
  • University of Bologna
  • University of Cagliari

DISI Sub-project details

Project astract The research project proposed by the research unit of Trento is aimed at developing and implementing adaptive procedures to<br/>manage the metrological confirmation state of sensor networks. Such a goal will be achieved by tackling the following issues:<br/>- Defining and analysing inductive procedures to check remotely the metrological confirmation state of sensor networks. These procedures should rely on the repetitive patterns and/or on the correlation between the information signals collected from multiple sensors and should be eventually implemented using programmable logic devices.<br/>- Analysing and characterizing from a metrological point of view multiple techniques in order to manage the periodic calibration of the whole measurement system;<br/>- Estimating the influence of measurement uncertainty on decision-making procedures.
Keywords Metrological confirmation, Learning systems, Programmable Logic Devices
Fundings 38300 €
Manager Dario Petri