Histogram and Other Aggregate Queries in Wireless Sensor NetworksReport as inadecuate




Histogram and Other Aggregate Queries in Wireless Sensor Networks - Download this document for free, or read online. Document in PDF available to download.

Histogram Queries, WSN, Median Queries, In-Network Query Processing, Database Systems

Additional contributors:

Subject-Keyword: Histogram Queries WSN Median Queries In-Network Query Processing Database Systems

Type of item: Computing Science Technical Report

Computing science technical report ID: TR11-03

Language: English

Place:

Time:

Description: Technical report TR11-03. Wireless Sensor Networks WSNs are typically used to collect values of some phenomena in a monitored area. In many applications, users are interested in summaries of the observed data, e.g., a histogram reflecting the distribution of the collected values. However, processing a histogram query efficiently on top of WSN is a topic that has not received much attention in the literature. In this paper we propose two main contributions: 1 an efficient algorithm for answering histogram queries in a WSN, and 2 how to efficiently use the obtained histogram to process other types of aggregate queries approximately and exactly.  Our experimental results using both synthetic and real data sets show that our proposed solutions are able to extend the lifespan of the WSN by up to one order of magnitude and at least double it.

Date created: 2011

DOI: doi:10.7939-R3NK3679Q

License information: Creative Commons Attribution 3.0 Unported

Rights:





Author: Ammar, Khaled Nascimento, Mario A

Source: https://era.library.ualberta.ca/


Teaser



1 Histogram and Other Aggregate Queries in Wireless Sensor Networks Khaled Ammar and Mario A.
Nascimento Department of Computing Science, University of Alberta, Edmonton, Canada { kammar,mn}@cs.ualberta.ca ✦ Abstract—Wireless Sensor Networks (WSNs) are typically used to collect values of some phenomena in a monitored area.
In many applications, users are interested in summaries of the observed data, e.g., a histogram reflecting the distribution of the collected values.
However, processing a histogram query efficiently on top of WSN is a topic that has not received much attention in the literature.
In this paper we propose two main contributions: (1) an efficient algorithm for answering histogram queries in a WSN, and (2) how to efficiently use the obtained histogram to process other types of aggregate queries approximately and exactly.
Our experimental results using both synthetic and real data sets show that our proposed solutions are able to extend the lifespan of the WSN by up to one order of magnitude and at least double it. 1 I NTRODUCTION A typical Wireless Sensor Network (WSN) consists of sensor nodes distributed in an area and connected, via a tree-like topology, to a base station.
Typically, WSN nodes have limited resources in terms of power, CPU and memory.
Battery lifetime is considered the most important resource in WSN nodes.
The required power for transmission is significantly higher than the required power for data processing in a WSN node.
For example, sending one bit using the Berkeley Mica motes needs as much energy as processing 1000 CPU instructions [9]. For this reason, it is very important that all algorithms run on top of a WSN minimize the transmission.
The base station is a full-fledged computer system with light limitations on memory, CPU, or bandwidth.
Nodes are typically used to observe some phenomena about a monitored area and are becoming common in many applications.
Examples include smart nursing, security monitoring, structu...





Related documents