Znanost razvoja i korištenja, praćenja i korištenja elektroničkih instrumenata u svrhu mjerenja, praćenja i zabilježavanja različitih fizikalnih fenomena koji mogu, a i ne moraju biti električke prirode. To uključuje analogne i digitalne elektroničke instrumente, sustave i standarde za mjerenje i bilježenje električkih veličina u frekvencijskoj i vremenskoj domeni, te mjerne pretvornike neelektričkih veličina. Predmet zanimanja su također i instrumenti s automatiziranim upravljanjem i analitičkim funkcijama.
Odjel za instrumentaciju i mjerenja
Zavod za elektroničke sustave i obradbu informacija i IEEE Instrumentation and Measurement Chapter pozivaju vas na predavanje:
"Energy Management in Wireless Healthcare Systems Using Dynamic Task Assignment"
Predavanje će održati prof. Tajana Šimunić Rosing, Department of Computer Science and Engineering, University of California, San Diego, na FER-u, u ponedjeljak 21. ožujka 2011. u 13:15 sati u Sivoj vijećnici. Predvidivo trajanje predavanja je 45 minuta.
Sažetak predavanja nalazi se u nastavku obavijesti.
Predavanje je otvoreno svima, a posebice pozivamo studentice i studente doktorskog i diplomskog studija.
Energy Management in Wireless Healthcare Systems Using Dynamic Task Assignment
Wireless sensor-based healthcare systems are hierarchical and heterogeneous in nature with components that have different energy and performance capabilities. In such systems with processing, sensing, and communication tasks, allocation of tasks to devices of the system affects the entire system battery lifetime and energy consumption.
In this work the problem of optimal task assignment with objectives relating to minimizing the total energy consumption and maximizing the system lifetime are first formulated using Integer Linear Program (ILP)-based solutions. The ILP based static solutions are able to improve the battery lifetime by up to 1.4 times in comparison to performing all of the processing tasks on the backend server. Given the dynamically changing nature of wireless systems and computationally expensive nature of ILPs, three heuristics are proposed. Dynamic task assignment algorithm, DynAGreen algorithm is a graph-based partitioning algorithm that assigns tasks on to system resources so that the total energy consumption of the system is minimum. DynALife algorithm is a heuristic task assignment strategy that extends system battery lifetime. DynAGreenLife has been designed to balance both system energy and system lifetime in wireless healthcare systems. All the three proposed dynamic algorithms are within 0.001% of the optimal ILP solution in static conditions. Our dynamic scheduling technique is able to improve system lifetime by up to 88% and on an average 30% in comparison to the static task assignment given by the ILP in dynamically changing urban conditions that represent real life scenarios.
