The Department of Computer Science at the University of Cyprus cordially invites you to the PhD Defense entitled:
KSpot+: A Network-aware Framework for Energy-Efficient Data Acquisition in Wireless Sensor Networks
Speaker: Panayiotis G. Andreou
Wireless Sensor Networks (WSNs) offer a non-intrusive technology that enables users to monitor the physical world at an extremely high fidelity. In order to collect the data generated by these tiny-scale devices, the data management community has proposed the utilization of declarative data-acquisition frameworks. While these frameworks have facilitated the energy-efficient retrieval of data from the physical environment, they were agnostic of the underlying network topology and also did not support advanced query processing semantics. In this dissertation we present KSpot+, a novel distributed network-aware framework for WSNs that optimizes network efficiency by combining three novel components: i) the Tree Balancing Module, which balances the workload incurred on each sensor node during a query by constructing efficient network topologies; ii) the Workload Balancing Module, which minimizes data reception inefficiencies by synchronizing the network activity intervals of each sensor node; and iii) the Query Processing Module, which provides advanced query processing semantics by employing a novel ranking mechanism that yields only the k-highest ranked answers, thus further minimizing energy consumption. The modules of the KSpot+ framework can operate both in isolation and in combination with each other, offering high degrees of energy efficiency, scalability and accuracy in the presence of failures. In order to validate the efficiency of our approach, we have created a prototype implementation of the KSpot+ framework in nesC and JAVA. In our experimental evaluation, we thoroughly assess the performance of KSpot+ using the real prototype system we developed and datasets from the University of California - Berkeley, the University of Washington and Intel Research Berkeley. We show that KSpot+ provides significant energy reductions under a variety of conditions, thus prolonging the longevity of a WSN as much as 317%, compared to predominant approaches.
Panayiotis G. Andreou is a PhD Candidate at the Department of Computer Science, University of Cyprus.
|Other Presentations Web: https://www.cs.ucy.ac.cy/colloquium/presentations.php|
|Colloquia Web: https://www.cs.ucy.ac.cy/colloquium/|