Energy-Efficient Cloud-Integrated Sensor Network Model Based on Data Forecasting Through ARIMA

Energy-Efficient Cloud-Integrated Sensor Network Model Based on Data Forecasting Through ARIMA

Kalyan Das (Department of Computer Science and Engineering, Veer Surendra Sai University of Technology, Burla, India) and Satyabrata Das (Department of Computer Science and Engineering, Veer Surendra Sai University of Technology, Burla, India)
Copyright: © 2022 |Pages: 17
DOI: 10.4018/IJeC.290292
OnDemand PDF Download:
No Current Special Offers


An energy-efficient Model for Sensor-cloud is proposed based on data forecasting through an autoregressive integrated moving average (ARIMA). Generally, all the user requests are redirected to the wireless sensor network (WSN) through the cloud. In the traditional approach, user requests are generated every fifteen minutes, so the sensor must send data to the cloud every fifteen minutes. In the current approach, the sensors within the WSN communicate with the cloud every two hours. The data forecasting technique addresses most of the user requests using the ARIMA one-step ahead forecasting model in the cloud. This results in less frequency of data communication, thereby increasing the battery life of the sensor. The ARIMA-based forecasting model provides better accuracy because of fewer temperature data changes with respect to the current temperature, for the next two hours. The proposed method for the simulation in the sensor cloud system consumes significantly less energy than the traditional approach, and the error in forecasting becomes highly negligible.
Article Preview


WSN has been used for various applications recently. Many applications like monitoring environments such as measuring temperature, humidity, speed of the wind, and rainfall need the help of WSN. Many physical sensors combine and transmit data wirelessly in the WSN. The cloud system provides storage, infrastructure, and other resources on rent to the users. Many organization uses cloud services to minimize the cost of buying sever and other resources like platform, software, and other services. The Sensor-cloud provides sensing service to the end-users using the cloud systems. The end users can use the sensor, which is attached in the cloud by the sensor owner. The sensor owner gets paid once the users utilize the sensor. By using the virtualization technique, one sensor can be accessed by multiple end-users. Sensor-cloud must utilize energy efficiency due to the limited lifetime of the battery in the sensor node. The cloud system consumes more energy for running the servers in the datacenter.

A sensor node S uses Pk amount of energy to transmit data to another sensor Skˊˊ with a rate of data transmission Rk. Here 1 ≤ kˊ, kˊˊ ≤ n, and kˊ≠ kˊˊ. Here Pk can be calculated (Guha et al., 2007) as follows:

(1) where:

P1 = the ideal power expenditure of sensor node SPˊ = constant Rk = the data rate of sensor SPrec = minimum energy required for successfully decoding at Skˊˊd = distance between the sensor nodes Skˊ and Skˊˊ

β lies between 2 and 6, depending upon the environment.

The consumption of energy in the WSN depends on the data rate and the distance between the nodes. More transmission of data can make the battery dry soon. Generally, user queries are generated fifteen minutes, so the sensor responds to the queries by sending data to the cloud every fifteen minutes. So forecasting schemes that forecast future sensor data for two hours in advance within the cloud system can save energy for the sensor network as sensors send data every two hours to the cloud system.

Complete Article List

Search this Journal:
Open Access Articles
Volume 18: 6 Issues (2022): 2 Released, 4 Forthcoming
Volume 17: 4 Issues (2021)
Volume 16: 4 Issues (2020)
Volume 15: 4 Issues (2019)
Volume 14: 4 Issues (2018)
Volume 13: 4 Issues (2017)
Volume 12: 4 Issues (2016)
Volume 11: 4 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing