Prediction Based Quantile Filter for Top-k Query Processing in Wireless Sensor Networks

Autor:	Baoli Song, Jiping Zheng, Haixiang Wang, Hui Zhang, Qiuting Han
Rok vydání:	2013
Předmět:	Base station Transmission (telecommunications) Computer science Sensor node Real-time computing Filter (signal processing) Autoregressive integrated moving average Wireless sensor network Energy (signal processing) Quantile
Zdroj:	Intelligent Computing Theories and Technology ISBN: 9783642394812 ICIC (2)
Popis:	Processing top-k queries in energy-efficient manner is an important topic in wireless sensor networks. It can keep sensor nodes from transmitting redundant data to base station by filtering methods utilizing thresholds on sensor nodes, which decreases the communication cost between the base station and sensor nodes. Quantiles installed on sensor nodes as thresholds can filter many unlikely top-k results from transmission for saving energy. However, existing quantile filter methods consume much energy when getting the thresholds. In this paper, we develop a new top-k query algorithm named QFBP which is to get thresholds by prediction. That is, QFBP algorithm predicts the next threshold on a sensor node based on historical information by A utoreg R essive I ntegrated M oving A verage models. By predicting using ARIMA time series models, QFBF can decrease the communication cost of maintaining thresholds. Experimental results show that our QFBP algorithm is more energy-efficient than existing quantile filter algorithms.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::cb72c30b69e9815bf3c0e801a86b4e8d https://doi.org/10.1007/978-3-642-39482-9_44 Zobrazit plný text záznamu