The Internet of Things introduces a new paradigm where small scale sensor devices can be used to capture data on physical phenomena. The potential scale of data that will be captured by such devices stands to transcend the capabilities of today’s client-server architectures. The necessity to reduce the data volume has been reﬂected in enormous research interests dedicated towards developing new data aggregation techniques for sensor networks. However, the application-speciﬁc nature of data aggregation techniques has resulted in the development of a large volume of options, thereby introducing new problems of appropriate selection. This paper introduces a unique method to deal with this problem by proposing a classiﬁcation approach for data aggregation techniques in wireless sensor networks. It presents the theoretical background for the selection of a set of high-level dimensions that can be used for this purpose, while a use case is presented to support the arguments. It also discusses how the dimensions dictate data collection procedures and presents how this framework can be used to develop an adaptive model for the dynamic selection of data aggregation techniques based on the characteristics of a sensing application use case.
|Title of host publication||Not Known|
|Publication status||E-pub ahead of print - 22 Nov 2018|
|Event||International Academic, Research, and Industry Association (IARIA) Data Analytics 2018 - Athens, Greece|
Duration: 18 Nov 2018 → 22 Nov 2018
|Conference||International Academic, Research, and Industry Association (IARIA) Data Analytics 2018|
|Period||18/11/18 → 22/11/18|
- Internet of Things
- Wireless Sensor Networks
- Big Data
- Data Aggregation
- Adaptive Model.
Omosebi, O., Bessis, N., Korkontzelos, Y., Pournaras, E., Sun, Q., & Sotiriadis, S. (2018). Dynamic Scenario-based Selection of Data Aggregation Techniques. In Not Known (pp. 27-32) https://www.iaria.org/conferences2018/DATAANALYTICS18.html