Distributed data clustering in multi-dimensional peer-to-peer networks
ADC '10 Proceedings of the Twenty-First Australasian Conference on Database Technologies - Volume 104
Peer-to-peer data mining classifiers for decentralized detection of network attacks
ADC '13 Proceedings of the Twenty-Fourth Australasian Database Conference - Volume 137
| Hi-index | 0.00 |
Sensor networks are usually composed by spatially distributed devices able to monitor physical or environmental conditions (pressure, temperature, motion, etc.). This kind of networks were originally built with the idea of transmitting elementary information to external sinks for further processing and querying. Nowadays, the growth of sensors memory and computational capability together with the significant reduction of energy consumptions have changing the potential of sensor networks allowing in-network storage and processing. The W-Grid infrastructure follows a Data Centric approach that indexes data according to any number of attributes so that it is possible to query events of interest through multi-dimensional range queries. Differently from existing Data Centric solutions W-Grid does not use either sensors physical position (i.e. GPS) nor estimation of their positions. For this reason W-Grid can be applied to a wider number of scenarios than existing solutions, as it works both indoor and outdoor, and can be easily suitable to other kind of ad-hoc networks, such as mesh networks and wireless community networks. In this paper we describe how W-Grid is able to efficiently managing and querying data in wireless sensor networks and we report, by means of an extensive number of simulations, several performance measures of its efficiency in comparison with a well-know competitor solution in literature.