Information Technology Reference
In-Depth Information
In the paper the authors show that the platform solves (through resources scal-
ability) the computational power requirements of environmental monitoring and
modeling applications.
Another work proposed by Ahmed and Gregory [4] presents an integration
framework between WSN and Cloud Computing. The main objective of the
proposed framework is to
“facilitate the shift of data from WSN to the Cloud
Computing environment
”. In addition, the authors suggest that the linkage of
Cloud Computing and WSNs allows the possibility of storage the WSNs data
in publics domains. Then, different users and applications can access to the
information of the sensors and these results in a better data usage.
Another platform to integrate WSN into Clouds is Aneka [5]. This platform
uses resources of private and public Clouds in order to provide support to ap-
plications of smart environments including health-care, transportation, urban
monitoring and others.
Regarding to the use of Clouds in agricultural environments, Hirafuji et. al. [6]
developed a Ambient Sensor Cloud System for High-throughput Phenotyping.
This platform allows the storage and access to data collected by sensor nodes
using Twitter Cloud services. The main goal of the system is to provide a simple
and economical solution to solve the access and storage of large datasets from
various sensor nodes. Hori et. al. [7] present a commercial solution to storage
and process WSNs data. The platform allows the integration with business man-
agement, production history, traceability and good agricultural practice systems
provided as a SaaS model.
Based on the works studied in this section, it can be concluded that Cloud
is a promising technology for solving the management and processing of data in
WSN's based AMS. Although most of the studied works use Amazon EC2, to
the best of our knowledge there are no works oriented to model the performance
and economic cost of EC2 instances in Agricultural Monitoring Systems.
4 Frost Prediction Application
In this Section we present the application for frost prediction. The main objective
of this application is to compute the minimum temperature that happen in the
night. Then, this temperature value is useful to predict if a frost may occur on
the farm. The Section is organized as follow: in subsection 4.1, we present the
method for frost prediction used in our application. Next, in subsection 4.2 the
application implementation is detailed.
4.1 Frost Prediction Method
The frost prediction application was developed using the frost prediction method
(FPM) of Snyder and de Melo-Abreu [15], which is based on Allen's equation [16].
The FPM predicts the minimum temperature that will occur in nights without
both clouds and cold fronts. Therefore, it is only suitable to predict radiation
frosts. The data used to calculate the minimum temperature are extracted from
