A Wireless Sensor Network System

For Monitoring Trees’ Health Related Parameters in a University Campus

Luis Eduardo Pérez, Jorge Arturo Pardiñas-Mir, Omar Guerra, Javier de la Mora, Mauricio Pimienta,

Nestor Hernández and Manuel de Atocha Lopez

Electronics, Systems and Informatics, ITESO University, Periférico Sur 8585, 45604, Tlaquepaque, Mexico

Keywords: Wireless Sensor Networks, Web Services, Mobile Application.

Abstract: This paper presents an experimental Wireless Sensors Network system which aims to contribute to the

transformation of a university campus into a living lab to experiment with applications in the context of a

smart city. The system acquires environmental data related to the campus trees’ health, stores it in a server

to be analysed and makes it available to be displayed in a mobile application. Details are given to the design

and results of the server’s system and the MobApp.

1 INTRODUCTION

Our cities around the world are having currently a

significant growth, the UN estimates that by year

2050 70% of the population will live in cities (UN,

2010). This will create significant pressure for cities

to efficiently provide services for the population:

water, energy, transportation, health care, education

and security. Cities must become "smart" (Naphade

et al., 2011) in how they manage their infrastructure

and resources to meet current and future needs.

The intelligence of cities is on the basis of

innovation and new working practices, primarily, on

the use of communication technologies and

information as a mean to achieve it. In this aspect,

the concept of all objects connected via internet

forms a technological basis for collecting

information from the city centers: water meters,

electricity meters, traffic sensors, parking meters,

temperature sensors, GPS devices, mobile phones,

etc. It is the concept called "Internet of Things" (or

IoT) (Foschini et al., 2011).

One of the key components of IoT is

undoubtedly wireless sensor networks of different

types (architecture, size, extent) which connect to

the Internet. They are able to provide the collected

information facing several challenges such as

holding systems’ owners and operational standards

that are not open, which prevents, for example, the

integration with other systems (Jiang et al., 2013).

We are developing a campus wide wireless

sensor network as an experimental platform to

research and implement solutions oriented to turn

the university campus into an intelligent community,

being a living lab to experiment with applications

that could be adapted in the context of a smart city.

One of such applications that we are using at the

same time as a demo of a living lab possibilities, as a

real laboratory for student lab work and research and

as a running service, is a system conceived to

collect, store and present information concerning the

health of the trees in the campus. This paper focuses

on the development of the computing and

information process stage of the application, leaving

for a later publication the details of the platform

hardware.

The system aims to collect information from a

network of wireless sensors making it available for

immediate query through a mobile application. At

the same time the information is stored in a web

server for later analysis. This represents the

integration of various emerging technologies; first of

all, the network of sensors reads temperature and

humidity data, secondly, the data is stored and

managed in a database on a server and, finally,

through a web service, a mobile application requests

the data from the server and presents it graphically.

It is expected that lab work from students and

research could be driven in these subjects under real

conditions. Some studies (Dospinescu, 2013),

(Moreira, 2011), (Choi, 2013), (Koo, 2011) which

refer to using a mechanism for sending / receiving

information through the use of web services using

REST and HTTP, are similar to our purpose.

Pérez L., Pardiñas-Mir J., Guerra O., de la Mora J., Pimienta M., Hernández N. and Lopez M..

A Wireless Sensor Network System - For Monitoring Trees’ Health Related Parameters in a University Campus.

DOI: 10.5220/0005546600420047

In Proceedings of the 12th International Conference on Wireless Information Networks and Systems (WINSYS-2015), pages 42-47

ISBN: 978-989-758-119-9

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

2 ELEMENTS OF THE SYSTEM

The university campus to be monitored has a surface

of over 41 hectares where 22 hectares correspond to

green areas. Here there are over three thousand trees

of over 280 different species. In order to maintain in

good health this valuable natural resource of the

university it is necessary to periodically monitor

some physical parameters that bring information

related to the conditions at which trees become

susceptible to contracting a disease or pest.

According to the requirements provided by the

office in charge of the gardens of the campus, the

parameters to be sensed in a first stage, are the

moisture and temperature of soil and the temperature

of trees.

The system that we are developing, Figure 1, has

already been concept tested in a first stage. It is

based on a wireless network of sensors distributed

all over the campus to collect the physical

parameters from the trees’ environment. This data is

transmitted to a station (server) in charge of storing

it and providing it to the users for its analysis. The

information is then looked up and displayed through

a mobile application via internet. An additional

feature is to show the route to go from any location

in the campus to a specific tree.

2.1 The Campus Wide Wireless Sensor

Network

At the base of the system there´s an experimental

wireless sensor network consisting of several sensor

nodes (SN). Each sensor node has the possibility of

sensing the values corresponding to the soil

moisture, humidity sensor (HS), and temperature of

soil (ST), and the temperature of trees (TT).

According to the requirements, some samples of

these values are sensed in a specific gardens’ area,

meaning that there’s no need of having a sensor

node for each and every tree of the campus.

The sensors are grouped into areas of trees

concentration and they communicate wirelessly to a

router node (RN). The router, one for each area,

manages the communication between sensors and a

coordinator node (CN) in charge of the whole

network. This coordinator node has also the

possibility of performing as a gateway to let the

network communicate with the system’s server by

internet. Figure 2 shows, as an example, a tested

WSN architecture comprising a coordinator, 3

routers and 3 sensor nodes per router. Each router

covers a specific tree-covered zone of the campus

where the monitoring is required. This architecture

is changing as we make the network grows and test

for the best results.

The WSN uses Zigbee technology to

communicate wirelessly between nodes while the

coordinator, at present, is connected to the internet

through an Ethernet port. We use XBee ZB RF

Modules from Digi International working at the ISM

2.4 GHz band. The low version is used in the sensor

nodes, running on batteries, while the PRO version

(more power) is used for the Router and the

Coordinator nodes running on electric power. A

more detailed description of the WSN is planned in a

coming publication.

2.2 The Server

The main functions of the systems’ server are: to

communicate with the WSN, to interact with the

mobile application (MobApp), and to manage the

database and to calculate the route to go from any

location in the campus to a specific tree.

The communication with the WSN is made with

the networks’ coordinator, through a TCP/IP

connection. It sends the required commands to read

the sensors’ values.

The interaction between server and the mobile

application uses web services based on REST and

SOAP. Here we are using SOAP as an educative

strategy for the students participating in the project.

The MobApp sends requests to the server for tree

information: species, georeference, planting date,

height, and health related variables (tree’s

temperature, soil’s temperature and soil’s moisture).

The sever has the possibility to respond with a XML

or JSON format file.

Figure 1: Components of the system.

AWirelessSensorNetworkSystem-ForMonitoringTrees'HealthRelatedParametersinaUniversityCampus

Figure 2: The Wireless Sensor Network Architecture.

The data base design takes into account an

existing old database of trees in the campus, still at

work, in order to normalize it and assure data

consistency. This data was linked to the sensors’

positions. In addition, a small program was made to

have the possibility of managing new entries and

relations between trees and sensors.

Due to the fact that main maps’ providers don’t

take into account the route path computation into

private spaces, like the University campus in this

case, it was necessary to develop in the server a web

service to compute the route to go from any point in

the campus to a targeted tree. For this means it was

necessary to consider the internal map of the

campus.

2.3 The Mobile Application

The Mobile Application was developed at the same

time for the iOS and Android platforms. Both of

them were designed with the same looking and

functionalities. The design process includes the

application requirements determination, the design

concept definition, and the user experience (UX) and

the user interface (UI) tests.

The entry data to the application is either the

number of a tree (all trees in the campus have a plate

with a number) or the name of a tree species. The

MobApp establishes an HTTP connection to send

the identifier. The server responds with the related

information from the database which is formatted in

one of two requested formats: XML or JSON. The

MobApp displays this information in a graphical

way. In the case of an inquiry using the tree’s

identifier the response includes all the tree’s

characteristics. For the case of an inquiry with a tree

species’ name the response is the map of the campus

showing the positions of all trees belonging to that

species. These results can be filtered for different

zones of the campus. Each tag on the map,

representing one tree, is sensitive to show the

information of that specific tree.

In the case of a route inquiry to go from any

location in the campus to a specific tree, the

MobApp sends to the server the user’s location and

the target tree’s location. The algorithm in the server

searches for the shortest route from the user to the

tree and sends the location of points of a valid path

to the MobApp in order to be displayed on the

campus map.

3 IMPLEMENTATION AND

RESULTS

The center of the system is a Linux server which

communicates with the WSN coordinator through a

TCP/IP connection. The communication program is

coded in Python, performing the necessary

commands to interact with the WSN protocol: IEEE

802.15.4. At this first stage of development we

acquire only the temperature and moisture of soil;

tree’s temperature as well as other environmental

variables will be added in a next stage.

The database was developed using MySQL

technology. It manages four kinds of information,

each one defined in a data table: information related

to the tree (rssy_arboles_inventario), information

related to the tree species

(rssy_arboles_taxnonomias), information related to

the garden zones (rssy_arboles_jardines) and

WINSYS2015-InternationalConferenceonWirelessInformationNetworksandSystems

information related to the sensors

(rssy_arboles_inspeccion).

The web services are accessed through an API

running in the server that was developed in PHP.

This API lets the server receive a request for a

specific tree or for a tree’s species. In the case of the

tree, the syntax needed to be used by the MobApp to

get the information in XML format is:

http://papvidadigital.com/risi_app/?

nid=ID

where ID is the tree identifier, while for getting the

JSON format, for example, the tree with identifier

99, is:

http://papvidadigital.com/risi_app/?

nid=99&format=json

The response received by the MobApp from the

server includes the whole information related to the

tree: identifier, species, planting date, diameter,

height, estimated health status (according to

temperature and moisture data), latitude, longitude,

garden’s zone, and image. Figure 3 shows the page

of the MobApp which displays the tree’s

information.

An example of a response for both XML and

JSON format are:

{“dato”:[{“dato”:{“NID”:“99”,”taxono

mia”:“Moringa”,“plantado”:“2010”,”di

ámetro”:”8,”altura”:”2”,”valoración”

:”100”,”latitud”:”20.606712”,”longit

ud”:”103.415931”,”jardin”:”1”,”imag

en”:””}}]}

In the case of a species request, the inquiry is

constructed from the tree identifier and the name of

the species, as shown here for both formats:

http://papvidadigital.com/risi_app/?

id_taxonomia=ID

http://papvidadigital.com/risi_app/?

id_taxonomia=2&format=json

The response, as said before, is the location of

every tree belonging to that species, as shown here

for the XML format case:

The process of showing the route from the user’s

location to the target tree was divided in two: first,

the path is calculated in the server after an inquiry

from the MobApp, and then, the server returns the

latitude and longitude corresponding to all points of

the path. Second, the MobApp takes these points and

shows them graphically on the campus map.

Figure 3: Mobile display of tree’s information.

AWirelessSensorNetworkSystem-ForMonitoringTrees'HealthRelatedParametersinaUniversityCampus

Figure 4: Flow diagram to compute the route from the user

to a target tree.

Figure 5: Display of the result of a route computing.

We programmed a simple algorithm for

calculating a path from the user to the target tree

following the possible real routes of the campus. For

this aim it was necessary to acquire and store in

advance the georeferenced data related to the

location points of campus’ valid walking routes. The

algorithm corresponding to the route computing is

shown in figure 4 while a display of a route result is

shown in figure 5. At this moment, the algorithm is

not yet optimal for analyzing all possible routes and

be able to know which the shortest one is.

4 CONCLUSIONS

We have presented a complete system consisting of

a wireless sensor network, a server provider of web

services, and a mobile application, that provides

information of some environmental parameters

related to the tree’s health in a university campus.

We have described details concerning the design of

the server and the MobApp based on the paradigm

of web services using REST and SOAP.

The project has a double objective. First, to

develop a system that provides a service for the

university. Second, to allow to experience with real

applications related to wireless sensor networks, web

services, and mobile applications. In this context, the

future work is first, to optimize the hardware and

management of the wireless network, second, to link

to the system a culture house belonging to the

university which is located at the center of the city

and, third, to develop an application that lets the

university’s office in charge of the campus’ gardens

to process and analyze the information.

All of this contributes to transform the campus

into a living lab for experimenting with applications

that could be in turn adapted in the context of a

smart city.

REFERENCES

Choi, M., Jeong, Y., Park, J., 2013. Improving

Performance through REST Open API Grouping for

Wireless Sensor Network. In International Journal of

Distributed Sensor Networks Volume 2013, Article ID

958241. http://dx.doi.org/10.1155/2013/958241

Dospinescu, O., and Perca, M., 2013. Web Services in

Mobile Applications. In Informatica Economica, vol.

17. DOI:10.12948/issn1431305/17.2.2013.02

Foschini, L.; Taleb, T.; Corradi, A.; Bottazzi, D., 2011.

M2M-based metropolitan platform for IMS-enabled

road traffic management in IoT. In Communications

Magazine, IEEE , vol.49, no.11, pp.50,57, November

2011. doi: 10.1109/MCOM.2011.6069709

Koo, B., Han, K., James, J., Shon, T., 2011. Design and

implementation of a wireless sensor network

architecture using smart mobile devices. In

Telecommun Syst (2013) 52:2311–2320. DOI

10.1007/s11235-011-9535-z

WINSYS2015-InternationalConferenceonWirelessInformationNetworksandSystems

Jiang, Y., Zhang, L., and Wang, L., 2013. Wireless Sensor

Networks and the Internet of Things. In International

Journal of Distributed Sensor Networks, vol. 2013,

Article ID 589750. doi:10.1155/2013/589750

Moreira, N., Venda, M., Silva, C., Marcelino, L., Pereira,

A., 2011. @Sensor – Mobile Application to Monitor a

WSN. In CISTI (Iberian Conference on Information

Systems & Technologies / Conferência Ibérica de

Sistemas e Tecnologias de Informação) Proceedings,

July.2011.

Naphade, M.; Banavar, G.; Harrison, C.; Paraszczak, J.;

Morris, R., 2011. Smarter Cities and Their Innovation

Challenges. In Computer , vol.44, no.6, pp.32,39, June

2011. doi: 10.1109/MC.2011.187

United Nations, 2010. World Urbanization Prospects: The

2009 Revision—Highlights, 2010; http://esa.un.org/

unpd/wup/doc_highlights.htm.

AWirelessSensorNetworkSystem-ForMonitoringTrees'HealthRelatedParametersinaUniversityCampus