A Metropolitan Area Living Lab based on a Wireless Sensor Network

Jorge Arturo Pardiñas-Mir, Luis Rizo-Dominguez, Luis Eduardo Pérez-Bernal,

Tino Hohler and Marino Esteban Pérez-Dorador

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

Keywords: Living Lab, Wireless Sensor Networks, Smart Cities, Environmental Monitoring.

Abstract: This paper presents a Living Lab based on a wireless sensor network with a metropolitan area dimension. It

is an experimental infrastructure providing real conditions to facilitate the development and testing of

technological solutions in the context of a smart city in subjects such as wireless sensor networks, wireless

data transmission, web services, software analytics and visualization systems. The first stage of the Lab is the

development of a base target application for sensing the environmental conditions at various locations in an

urban area.

1 INTRODUCTION

The concept of Living Lab has many meanings, being

the most mentioned the one which states that is a

living environment which houses both people and

technology, in a semi experimental setting that

promotes symbiotic innovation, development and

research (Chin and Callaghan, 2013). This

signification is sometimes extended or limited

according to the vision behind a particular

application. (Del Vecchio et al., 2014) presents a

review of the literature and contributions about the

Living Labs focusing on their implications for the

development of entrepreneurial competencies. In the

field of education, for example, (Justice and Do,

2012) presents a living lab to promote learning

through challenging real-life hands-on experiences

that are supervised by faculty, students and staff. In

(Vicini et al., 2012) the Living Lab design approach

is applied into a paediatric section of a hospital to the

understanding, studying and measuring of the

interaction between children and services and the

potential of Internet of Things in innovation. This

paper defines the Living Lab as an experimental

infrastructure with an urban area dimension,

providing real conditions to facilitate the

development and testing of technological solutions in

the context of a smart city. The technical contribution

of this work is to bring to the entrepreneurial and

educational community a platform that allows quick

access to the validation of the concept of a new

product, to generate experience, and eventually

identifying new products to strategic markets in five

main lines of technologies: wireless sensor networks,

wireless data transmission, web services, software

analytics and visualization systems.

The design and implementation of the proposed

system has as base target application the sensing of

the environmental conditions at various points in an

urban area. This will let characterize the air quality in

the zone, generating information that can be exploited

and analysed by additional software tools. The main

features of the system defined for the first stage were

the use of wireless technology for interconnection,

being hybrid in such a way that it allows the

coexistence of different wireless technologies, an

open architecture that allows the system to easily

expand and improve, the ability to add additional

sensors, and its ease of being replicated. The system

built for this application stays in a stage ready to be

used for other kind of applications, starting to play its

role as a Living Lab. It was already being used into

an Internet of Things university course.

Section 2 presents the context of the application

for the first stage of the Living Lab. Section 3

describes the elements of the system while Section 4

shows the results of the first stage of deployment.

Finally, some conclusions and a discussion of the

future work are introduced.

158

Pardiñas-Mir, J., Rizo-Dominguez, L., Pérez-Bernal, L., Hohler, T. and Pérez-Dorador, M.

A Metropolitan Area Living Lab based on a Wireless Sensor Network.

DOI: 10.5220/0005970501580164

In Proceedings of the 13th International Joint Conference on e-Business and Telecommunications (ICETE 2016) - Volume 6: WINSYS, pages 158-164

ISBN: 978-989-758-196-0

2 MONITORING

ENVIRONMENTAL

CONDITIONS

There are two application goals for the Living Lab at

this first stage. One is to have a computational tool for

acquiring information to help the forecasting of

environmental contingency situations, for example,

to prevent when a pollutant could exceed its

allowable limit. The second, is the monitoring of

environmental conditions around the most important

forest in the city. This serves to identify the way and

speed in which the forest is degraded because of

urban growth and to help taking corresponding

actions to support the survival of the forest.

According to international standards (EPA, 2013), it

was decided to monitor the so called criteria

pollutants: the ozone, the carbon monoxide, the

sulphur dioxide, the nitrogen dioxide and particulates

less than 10 microns (PM10). Unlike employ a

weather station for measuring ambient air pollutants,

as is usually done, the use of wireless sensors can

provide a more practical, economic and smaller

solution at the possible expense of the measurement

accuracy.

To identify the technological possibilities used in

this kind of applications, some examples reported in

the literature can be mentioned. A system measuring

carbon monoxide and fine particles is reported in

(Wang et al., 2012) and (Liu et al., 2012). It is based

on an electronic card using a low power consumption

microcontroller from Texas Instruments and a ZigBee

radio circuit. The sensors are solid state and low cost.

The wireless network is controlled by a Gateway

based on an industrial personal computer. A different

approach is proposed in (Devarakonda et al., 2013),

where a mobile sensor sends the detected values

wirelessly by a cell phone linked to a server in

internet. This makes the information available on the

web. Here two kind of nodes are proposed, one to be

placed on a vehicle and one to be carried by a person.

The first one is built based on an Arduino card and a

cellular communication shield with both a fine

particles and a carbon monoxide sensors. The

personal node uses a commercial device named

NODE having the ability to measure carbon

monoxide, moisture, temperature, atmospheric

pressure and ambient light. Finally, a different

approach is presented in (Boubrima et al., 2015),

where a model is described to position environmental

pollution sensors in a city in order to lower the cost of

implementation, being the number of sensors the

main objective.

Figure 1: General view of the proposed system.

3 SYSTEM OVERVIEW AND

DESCRIPTION

As mentioned before, the proposed system is a

wireless network of sensor nodes (SN) dedicated to

measure the air quality at different places of an urban

area, as shown in Figure 1. The collected information

is stored on a server on the internet and made

available for consultation. Similarly, a website and a

mobile application allow access to the information.

3.1 General Structure and Operating

Principle

The structure of the system consists of the following

elements, illustrated in Figure 2.

 Several sensor nodes. Each node with the sensors

to measure the pollutants. They can be connected

to the internet through different kind of wireless

links;

 A server on the Internet. It keeps track of each

node and stores the collected information in a

database. It also provides web services necessary

for the operation of the website and the mobile

app;

 A web application that allows system

management. To add and delete nodes, to add and

delete sensor variables to be measured, and to

assign a location to each node;

 A Web page and a mobile application. They show

the position of the sensor nodes on a map and

provides current and historical values of the

sensors.

First, the way in which the system operates was

defined, establishing the main characteristics of each

A Metropolitan Area Living Lab based on a Wireless Sensor Network

159

element of the system. The main operation of the

system is based on the communication between the

sensor nodes and the server. For this an own protocol

for messages exchange was defined. Starting from the

normal state of the node, low power consumption, it

periodically awakes and perform some tasks such as

to take samples of the variables to be measured, send

them to the server, and ask the server for a specific

action to be done. The server is continuously listening

for node messages and eventually can ask the sensor

node for some action.

For the sensor nodes we used off-the-shelf devices

that we specifically programmed for the system, for

which we created some particular libraries. We

defined a communication protocol between the nodes

and the server as well as a gateway to cope with the

ZigBee technology. We developed all the needed

software, web services, web pages, and the mobile

applications. In the following subsections we describe

the characteristics of each element and some of the

challenges for its realization.

3.2 The Sensor Node

The sensor node is the starting point of the system. In

order to maintain its electrical autonomy its operation

was defined to remain in a low power state most of

the time and periodically awake and perform the

following tasks: to take samples of the variables to be

measured, to store locally those values, and connect

to the server and perform requested actions such as

data transfer to the server, changing settings, etc.

The choice of the technology to be employed for

the sensor nodes took into account the following

requirements:

 Ability to quickly deploy a solution;

 Being easily configurable or programmable;

 Having a set of robust basic libraries ready to be

used in different applications;

 Ability to enhance its hardware platform;

 Interoperability of wireless technologies.

We analysed four possible options that could meet the

requirements:

 Fully design the sensor node;

 Use of general purpose electronic control devices

such as Arduino and Raspberry Pi;

 General purpose off-the-shell wireless sensor

devices, like Digi’s XBee and Memsic’s sensor

nodes;

 Specific Application, open architecture, wireless

sensor nodes, like Libelium’s Waspmote.

We found that the option that best fulfilled the requi-

rements was the Waspmote from Libelium, a Spanish

company mainly dedicated to wireless sensor

networks products (Libelium, 2016). Either way,

being a device with an open architecture, the

development of the system took into account the

possibility of using any other technology and not

staying constrained to the use of only one provider.

These Waspmotes are based on a controller card

using an ATmega1281 microcontroller, a real time

clock and a solar cell rechargeable battery controller.

The sensor interfacing is done through special adapter

cards with the possibility of using user made

customized cards. It exists two versions of the

Waspmote, one aimed to experiment and develop

OEM solutions, and the other called Plug&Sense

aimed to quickly deploy the device for a specific

application, they are shown in Figure 3.

Figure 2: Elements of the wireless sensor network system.

Figure 3: The Waspmote OEM version, left, and the

Plug&Sense! Version, right.

The communication between the nodes and the

network for the first stage of the project follows two

connection strategies related to two different

application scenarios. The one without a near

connection to the internet, where a link through a

ZigBee network from the sensor node to a gateway

WINSYS 2016 - International Conference on Wireless Networks and Mobile Systems

160

with internet connection is used. The second scenario

envisages having a close internet connection, where a

link through a wireless local area network can be

established. We also considered and tested a GPRS

link, but at the time of writing this document we still

haven´t deployed a sensor node with such a

communication technology. In both application

scenarios it is necessary to have a Gateway that can

receive signals from the node and send them over the

internet to the server. This situation is shown in

Figure 4. For the WiFi case a commercial wireless

router was used, while in the ZigBee case a

customized Gateway was built.

Figure 4: Communication between server and nodes

through a Gateway.

3.3 The Server

The server consists of a set of web applications and

web services that perform as a whole the following

functions:

 To periodically request to the sensor nodes the

data collected and stored in its local memory;

 To make available the database on the Internet;

 To provide information related to the nodes

connected to the system, such as geographical

position and sensor variables that can be

measured;

 To provide information to the user related to the

data being collected from the system;

 To provide an interface for managing nodes (add,

delete and modify) and communicate with them

(action messages).

As mentioned before, a proprietary protocol for

exchanging messages between the sensor nodes and

the server was defined. The main command is the

Data Request (DR) and Transmit Data (TD) pair, for

which the node prepares the information and sends it

to the server. Figure 5 shows the flow of messages

between the sensor node and the server for this

function in the particular case of using a ZigBee link.

In this case, because of limits in length of a ZigBee

frame, the information from the sensor node must be

divided in smaller packets in order to be managed by

the ZigBee gateway which repack them and send

them to the server as a one whole frame.

At the beginning, the sensor node awakes and

sends a message request (MR) to the server through

the ZigBee gateway. The gateway passes the message

directly to the server. The server responds with a Data

Request (DR), message that is directly passed from

the gateway to the node. The sensor node prepares the

data and because its length is longer than a standard

ZigBee frame, it is divided in 2 packets for the case

illustrated in Figure 5. The gateway collects the two

packets and builds a single frame to send to the server,

who finally responds confirming the good reception

Figure 5: Time chart example of the communication protocol.

A Metropolitan Area Living Lab based on a Wireless Sensor Network

161

Figure 6: Screenshots of the mobile application.

of the frame with a Correct message.

This way of interacting between the gateway and

the node allows to build standard frames to be sent to

the server, avoiding the use of an exclusive treatment

of data for the ZigBee case at the server.

3.4 System Management Web

Application

The system management is made through a web

application developed in php and running in the

server. It allows to register, unregister, and edit the

sensor nodes and its characteristics, and to make the

same thing with the sensor to be employed. For each

node you can fix its protocol identifier, its position,

and the different sensors associated.

For each sensor you can fix the label to be used in

the protocol, and the name and units to be shown in

the information display application. According to

this, the system allows you to add any new sensor to

be measured and to associate any sensor to a

particular node independently of the rest of nodes and

sensors.

This gives the system the versatility of not being

constrained to only one kind of application.

In addition to the functions above, the

management application can generate a set of basic

data reports, either by node or by dates.

3.5 Website and Mobile Application

The website and the mobile application were

developed to show the information collected by the

system in a simple way. The first one is optimized for

a computer screen, while the second one is optimized

for a mobile device.

The function assigned to these elements is to

display the following information:

 The position of the sensors on a map;

 The sensors associated to a node;

 The latest value of a chosen sensor;

 The historical values of a chosen sensor.

Figure 6 illustrates the appearance of the mobile

application.

4 RESULTS

As the time of writing this paper, 4 sensor nodes have

been deployed along the city. Figure 7 shows one of

the nodes installed on a roof. They measure 4 of the

criteria pollutants: ozone, carbon monoxide, carbon

dioxide, and nitrogen dioxide, plus air temperature

and relative humidity.

Figure 7: A sensor node located at the roof of the

department building.

WINSYS 2016 - International Conference on Wireless Networks and Mobile Systems

162

Three of the deployed nodes are connected to the

internet via WiFi while the fourth one is connected by

a ZigBee gateway. The ZigBee connection is made up

with XBee ZigBee S2 radio modules from Digi

International. Each node has been powered with a 3.7

V - 6600 mAh battery rechargeable through a 7V –

500 mA solar cell.

The debugging of the node and the server program

was made possible through the use of two specially

developed web services intended to closely follow the

different stages of the system. A log service shows all

messages exchanged between the server and the

nodes, and a read service lists a number of latest

information frames being stored in the server. Table

1 shows an example of the data corresponding to one

information frame.

In the first stage of the project it was chosen to use

standard sensors with a medium accuracy. The values

collected will be analysed and compared to reference

sensors of a higher accuracy in order to adequately

calibrate the standard ones. Figure 8 shows a screen

corresponding to the historical values of one of the

sensors measuring the temperature of the

environment.

Table 1: Sample of registered data.

id id Node

17606 A1

data

{"id":"A1","ac":"td","ts":"2016-03-28T18:06:14-

0400","bat":"97","temp":"31.61","hum":"6.79","c02":"2.72","

no2":"0.14","03":"0.00","co":"0.87"}

date_hour

2016-03-28 18:06:14

Figure 8: One month historical values graph for the

temperature of the environment.

The performance of the electrical autonomy of the

nodes was monitored through the charge capacity of

the batteries. It was found that the nodes rest

autonomously powered by the rechargeable battery

and the solar cell. This is the case for two of the nodes

which have lasted for more than 5 months without

additional powering. The other two nodes were

powered recently. Figure 9 shows the average charge

percentage of the battery of one of the nodes for the

last 5 months. The performance of the charge capacity

of the battery without recharging was also tested.

Figure 10 shows that after almost two weeks of

operating without solar cell the charge of the battery

dropped from 97 % to near 80 %, meaning that during

a similar period the node can survive before a new

recharge in case of a failure of the solar cell.

Figure 9: Average charge percentage graph of a node’s

battery with solar cell during 5 months.

Figure 10: Average charge percentage graph of a node’s

battery without solar cell during twelve days.

5 CONCLUSIONS AND FUTURE

WORK

This paper presents the development and some

experiences regarding the implementations of a

whole system for monitoring the air quality at

different locations in a city through a wireless sensor

network and an internet based information and

management system. All of this working under real

conditions. It represents the materialization of an

infrastructure to experience and validate tangible

A Metropolitan Area Living Lab based on a Wireless Sensor Network

163

concepts and proposals to develop solutions to the

problems present in cities, a Living Lab for smart

cities.

In the first stage of development of the system 4

sensor nodes have been deployed and its good

operation has been verified.

In the next stage of development the calibration of

the standard sensors employed in the system will be

adjusted compared to factory calibrated accurate

sensors. New sensor nodes will be added to the

system using new communication links as 3G and

LoRa.

The use of the system as a Living Lab has already

started into an undergraduate course about wireless

sensor networks and Internet of Things. It has also

began the collaboration with a research group

dedicated to strength a buffer and transition zone

around the most important forest near the city aimed

to the conservation of the forest. The location of next

nodes will be set according to the objectives of the

group.

ACKNOWLEDGEMENTS

This work was supported by CONACYT and Hewlett

Packard Centro de Servicios Globales S. de R.L. de

C.V. through ProInnova 2015 funds.

REFERENCES

Chin, J., Callaghan, V., 2013. Educational Living Labs: A

Novel Internet-of-Things Based Approach to Teaching

and Research. In 2013 9th International Conference on

Intelligent Environments, Athens.

Del Vecchio, P., Elia, G., Ndou, V., Secundo, G., Specchia,

F., 2014. To what extent the practice on living labs

match with the theoretical framework? The case of

VINCENTE Living Lab for the creation of technology

entrepreneurship. In 2014 International ICE

Conference on Engineering, Technology and

Innovation (ICE), Bergamo.

Justice, C., Do, L., 2012. IT experiential learning: The

Living Lab. In Frontiers in Education Conference

(FIE), Seattle, WA.

Vicini, S., Bellini, S., Rosi, Sanna, A., 2012. An internet of

things enabled interactive totem for children in a living

lab setting. In 2012 18th International ICE Conference

on Engineering, Technology and Innovation (ICE),

Munich.

United States Environmental Protection Agency (EPA),

2013. Quality Assurance Handbook for Air Pollution

Measurement Systems, Volume II, Ambient Air Quality

Monitoring Program, EPA-454/B-13-03.

Wang, Ch., Huang, Y., Zheng, X., Lin, T., Chuang, Ch.,

Jiang, J., 2012. A self-sustainable air quality monitoring

system using WSN. In 2012 5th IEEE International

Conference on Service-Oriented Computing and

Applications (SOCA).

Liu, J., Chen, Y., Lin, T., Lai, D., Wen, T., Sun, Ch., Juang,

J., Jiang, J., 2011. Developed urban air quality

monitoring system based on wireless sensor networks.

In 2011 Fifth International Conference on Sensing

Technology (ICST).

Devarakonda, S., Sevusu, P., Liu, H., Liu, R., Iftode, L.,

Nath, B., 2013. Real-time air quality monitoring

through mobile sensing in metropolitan areas. In

Proceedings of the 2nd ACM SIGKDD International

Workshop on Urban Computing (UrbComp '13). ACM,

New York, NY, USA.

Boubrima, A., Matigot, F., Bechkit, W., Rivano, H., Ruas,

A., 2015. Optimal Deployment of Wireless Sensor

Networks for Air Pollution Monitoring. In 2015 24th

International Conference on Computer Communication

and Networks (ICCCN).

Libelium Comunicaciones Distribuídas, 2016.

http://www.libelium.com/ [Accessed February 26,

2016].

WINSYS 2016 - International Conference on Wireless Networks and Mobile Systems

164