Intelligent Acquisition Techniques for Sensor Network Data

Intelligent Acquisition Techniques for Sensor Network Data

Elena Baralis (Politecnico di Torino, Italy), Tania Cerquitelli (Politecnico di Torino, Italy) and Vincenzo D’Elia (Politecnico di Torino, Italy)
DOI: 10.4018/978-1-60566-328-9.ch008
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

After the metaphor “the sensor network is a database,” wireless sensor networks have become an important research topic in the database research community. Sensing technologies have developed new smart wireless devices which integrate sensing, processing, storage and communication capabilities. Smart sensors can programmatically measure physical quantities, perform simple computations, store, receive and transmit data. Querying the network entails the (frequent) acquisition of the appropriate sensor measurements. Since sensors are battery-powered and communication is the main source of power consumption, an important issue in this context is energy saving during data collection. This chapter thoroughly describes different clustering algorithms to efficiently discover spatial and temporal correlation among sensors and sensor readings. Discovered correlations allow the selection of a subset of good quality representatives of the whole network. Rather than directly querying all network nodes, only the representative sensors are queried to reduce the communication, computation and power consumption costs. Experiments with different clustering algorithms show the adaptability and the effectiveness of the proposed approach.
Chapter Preview
Top

Sensor Network Applications

Nowadays wireless sensor networks are being used for a fast-growing number of different application fields, with varying functional and operational requirements. Sensor network applications can be classified into two main classes: Habitat monitoring (Szewczyk et al., 2004) and surveillance applications (He et al., 2004). The habitat-monitoring applications (e.g., environment monitoring, highway traffic monitoring, habitat monitoring (http://www.greatduckisland.net/)) continuously monitor a given environment, while surveillance applications (e.g., health care monitoring, avalanche detection, condition maintenance in industrial plants and process compliance in food and drug manufacturing (Abadi et al., 2005)) alert the control system when a critical event occurs in an hostile environment or context. In the last case the alert needs to be detected with high confidence and as quickly as possible to allow the system to react to the situation. Some sensor network applications are described in the following.

Complete Chapter List

Search this Book:
Reset