A WEB OF THINGS (WOT) APPROACH TO SMART HOUSEHOLD

ENERGY MANAGEMENT FOR SUSTAINABLE LIVING

Sita Ramakrishnan and Subramania Ramakrishnan

Clayton School of IT, Monash University, Victoria, Australia

Keywords:

Web of Things, Smart Appliances, Internet of Things, Energy Consumption, Sustainability, 6LoWPAN/IPv6,

REST.

Abstract:

With a focus on a software system that incorporates smart phones, web-enabled physical devices and RESTful

APIs, this paper explores a system strategy for monitoring, integrating and controlling electrical devices to

facilitate the management of electrical energy consumption in line with modern sustainability practices. Real-

time feedback on energy consumption and associated costs for individual appliances in a household is the

key to helping the consumers in their quest for sustainable living. The paper considers a case study of a

household in Australia, having a grid-connected solar panel installed for electricity generation. The focus of

the case study is dynamic adaptation of both grid-supplied and the solar-generated electricity for powering

heavy house hold electrical loads with a view to reduce costs and greenhouse emissions.

1 INTRODUCTION

Modern ICT developments enable physical devices

manage their interaction with less human interven-

tion in order to avoid human bottleneck, and be more

autonomous. However there are many difﬁculties in

current internet architecture to accommodate various

physical data sources, actuators and distributed com-

puting elements. Autonomous networks of embedded

devices are centered around data fusion which is the

result of integrating a number of embedded devices

such as sensors with the internet (Abdelzaher, 2006).

Unlike traditional embedded systems, where the em-

phasis is more on computational elements and less on

the links between the physical and computational el-

ements, a cyber-physical system is designed as a net-

work of interacting elements with physical input and

output (Abdelzaher, 2006).

This paper considers Web of Things (WoT) and

cyber-physical systems with a case study scenario for

managing energy consumption to reduce costs and

greenhouse emissions in a household by adapting the

usage of electrical devices in the household. A lay-

ered architecture to address the distributed nature of

embedded, autonomous cyber-physical systems and

modelling cyber and physical resources in a uniﬁed

framework are considered in this paper.

Increases in recent times in costs of electricity,

gas, water etc., and the associated increases in envi-

ronmental pollution, such as greenhouse emissions,

are having an impact on lifestyles of people to adopt

lifestyle strategies for reducing cost and environmen-

tal pollution. Real-time feedback on energy consump-

tion and associated costs for individual appliances in

an easily readable format is the key to helping con-

sumers in changing their behaviour in their quest for

sustainable living. Merging physical household de-

vices with computing such as with embedded micro-

processors and wireless communication is leading to

the creation of so-called smart appliances. We explore

the way in which these smart appliances can be ad-

ministered through mobile web to facilitate the vision

of web-enabled smart home for sustainable living.

The paper is structured as follows. The next sec-

tion discusses advances in internet protocols and var-

ious home automation standards. Also discussed is

the ability of 6LoWPAN standard to make the lat-

est Internet protocol (IPv6) available to even the most

minimal embedded devices and its suitability for em-

bedding into household appliances. A system archi-

tecture for household smart appliances is then derived

by augmenting accepted protocols of WoT and an ex-

tended WoT framework after a discussion on Internet

of Things (IoT) and Web of Things (WoT). In Section

3, home requirements for managing electricity con-

sumption are described. In Section 4, a case study is

presented for a smart household energy management

for sustainable living. Section 5 provides conclusions

Ramakrishnan S. and Ramakrishnan S..

A WEB OF THINGS (WOT) APPROACH TO SMART HOUSEHOLD ENERGY MANAGEMENT FOR SUSTAINABLE LIVING.

DOI: 10.5220/0003945900290036

In Proceedings of the 2nd International Conference on Pervasive Embedded Computing and Communication Systems (PECCS-2012), pages 29-36

ISBN: 978-989-8565-00-6

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

and possible future extensions.

2 WOT FOR SMART

APPLIANCES

2.1 Internet Protocols and Home

Automation Standards

The internet has been a great success over two

decades with ubiquitous use of the network by bil-

lions of people. The internet paradigm has been

very successful with heterogeneous networks, and the

www model of uniform resource locators (URLs),

the hypertext transfer protocol (HTTP) and hypertext

markup language (HTML). This has resulted in IT

architects, communication experts and others to in-

novate and add new protocols and uses for internet

technology. Another internet advance has been the

internet of things where the embedded devices, also

named, smart objects, are universally becoming IP-

enabled and an integral part of the internet. Examples

of IP-enabled embedded devices are mobile devices,

personal health assistance devices, home and indus-

trial automation, smart metering etc (Guinard, 2010;

Kamilaris et al., 2011).

There are many competing home automation stan-

dards such as the popular X10 in the residential

market since 1978 (www.x10.com), ZigBee standard

based on IEEE 802.15.4, and KNX, the European

standard. Currently, monitoring and controlling em-

bedded devices are predominantly done using ser-

vices built on internet technology. Internet of things

is a powerful paradigm which has combined the

internet-enabled embedded devices and web services

technology. Till recently, the complexity of commu-

nication standards, protocols and services meant that

internet enabling happened only with the most pow-

erful embedded devices.

The ﬁrst global low-power radio standard was re-

leased by IEEE, which was the 802.15.4 low-power

wireless personal area network (WPAN) standard in

2003. A new paradigm was required to enable low-

power wireless devices with limited processing capa-

bilities to participate in the Internet of Things (Hui

and Culler, 2008; Shelby and Bormann, 2010). The

6LoWPAN standard makes this latest Internet proto-

col (IPv6) available to even the most minimal embed-

ded devices over low-rate wireless networks (Design,

2011), and is well suited for embedding into home

appliances.

Since typical households have cable Ethernet con-

nections and Wi-Fi as the backbone network, exten-

sion via 6LoWPAN is possible without laying addi-

tional cables. Although deployment costs for both

ZigBee and 6LoWPAN (IPv6) are low, ZigBee being

a proprietary solution requires its own infrastructure

as opposed to using LANs for 6LoWPAN/IPv6 [24].

Seamless integration to the internet is the attractive

feature available with IPv6 over the other standards.

IPv6 connectivity of smart appliances integrates these

appliances to the internet network layer. Web servers

running on the appliances are required to achieve web

integration (application layer). The concept of Re-

source is a ﬁrst class object in the REST (represen-

tational state transfer) style and every resource must

understand the core operations (Richardson and Ruby,

2007). A smart appliance resource is accessed in a

web-oriented manner using a lightweight REST ar-

chitectural style. Any smart appliance resource is

bound to a Uniform Resource Identiﬁer (URI), which

identiﬁes the resource involved in an interaction be-

tween entities. It uses HTTP 1.1 as a true applica-

tion protocol and its operations, GET, PUT, POST,

DELETE. Each resource implements a set of these

well-deﬁned operations. Applications using web-

enabled appliances can be developed using web lan-

guages such as Javascript, PHP, JSON (javascript ob-

ject notation) and toolkits such as JQuery. Android

and IPhone OS 4.0 run IPv6.

It follows from the discussion in this section that

6LoWPAN /IPv6 is suitable for smart home automa-

tion applications.

2.2 IoT, WoT and WoT Framework

The web is a very effective, user-centric, scalable, dis-

tributed platform with underlying technologies such

as TCP/IP, HTTP, HTML/XML, JSON etc. This suc-

cess has now been extended to incorporating real-

world objects into WWW using web technologies,

and this is called the Web of Things (WoT), as de-

scribed next.

The Internet of things (IoT) (Papadimitriou, 2009)

is about principles and technologies that enable the

internet to get into the real-world of physical objects.

IoT gives every device an IP address and lets it plug

into the internet. In IoT, everyday devices and objects

(objects that contain an embedded device) are con-

nected by integrating these into the web. Examples

of smart devices are sensor network, household appli-

ances etc. Web of Things (WoT) is an extension to

IoT (Zeng et al., 2011) and is about reusing the web

standards and building on the success of web 2.0.

Well-established web standards and blueprints

such as HTTP, REST, URI etc are used in WoT. This

ensures ease of development with existing web frame-

PECCS 2012 - International Conference on Pervasive and Embedded Computing and Communication Systems

works in accessing the functionality of smart objects.

In WoT, real-world objects such as consumer appli-

ances are integrated into the WWW by representing

them as web resources, which can be accessed us-

ing lightweight APIs based on REST principle (Oster-

maier et al., 2010). In WoT, real-world objects includ-

ing their sensors and actuators are exposed as URLS.

End users are able to create physical mash-ups by

composing personalized services based on physical

resources. A simple WoT framework is proposed for

our case study for home automation where the com-

ponents are distributed among the devices and could

also be in the supplier grid server (e.g. electricity

company).

In WoT, HTTP is used as an application protocol

rather than as a transport protocol, and the blueprint of

Resource-oriented architecture (Guinard et al., 2010;

Guinard, 2010) is followed for exposing the syn-

chronous functionality of smart objects through a

REST interface, known as Restful API. So, with WoT,

one should be able to use universally accepted pro-

tocols to connect a number of physical objects in a

loosely coupled, scalable manner.

A WoT framework to augment the accepted pro-

tocols to deal with some of the speciﬁc requirements

of cyber-physical systems such as information-centric

protocol, context-awareness, deterministic QOS etc

is necessary to address the grand challenge of cyber-

physical systems as envisaged by NSF (Dillon et al.,

2011). Developers would beneﬁt from a WoT frame-

work that can hide low-level implementation de-

tails and provide an application development environ-

ment for faster system development (Kamilaris et al.,

2011).

The WoT framework proposed by Dillon et al.

(Dillon et al., 2011) consists of a number of layers

from the physical device layer to program interface

layer (Caporuscio et al., 2011). The layers are: WoT

device, WoT kernel, WoT overlay network, WoT con-

text and WoT API. The WoT framework is above the

physical interface such as sensors, actuators, which

interact with the physical environment (Dillon et al.,

2011). A WoT framework allows the web world to

control the physical world using the data to perform

smart tasks such as smart home automation for sus-

tainable living, factory automation etc. (Kamilaris

et al., 2011).

WoT device provides a resource based abstraction

for the devices. Each physical device is modeled as a

WoT resource that has a universal identiﬁer, name and

a state. The pervasive REST architectural style pro-

motes the use of Resource as a ﬁrst-class object and

entities are modeled as resources, which can act both

as clients and servers (Erenkrantz, 2009; Caporuscio

et al., 2011). WoT device interacts with smart physi-

cal appliances (resources) through their interfaces.

WoT kernel provides a low-level run-time for

communication and management of WoT resources.

WoT kernel uses the notion of continuation that has

been used in AJAX and Mashups. It is responsible

for detecting newly connected or disconnected physi-

cal devices and their resources. WoT overlay network

provides a network aware logical abstraction on top of

the current internet architecture such as TCP/IP. WoT

context discovers and constructs contextual informa-

tion from the event stream in the overlay. WoT API

provides abstractions that allow developers to inter-

act with the WoT framework. WoT resource is the

abstraction unit. Application developers can interact

with each WoT device.

2.3 System Architecture

A system architecture for a Web of Things approach

to smart household energy management for sustain-

able living is proposed in this paper by augmenting

the accepted protocols of WoT and the extended WoT

framework discussed above. The energy monitoring

and consumption system is made up of the household

energy appliances and devices with sensors for mon-

itoring the energy data from various sources, a gate-

way software layer with a web server and database for

manipulating and storing each smart appliance’s en-

ergy consumption values in a lightweight manner, and

a presentation layer with a mobile phone for showing

the various energy consumption views of the appli-

ances as per customer requirement. This four layer

architecture (see Figure 2) is expanded with a case

study scenario (see Section 4).

3 MANAGING CONSUMPTION

Figure 1 shows a typical house in Melbourne, Vic-

toria, Australia, that has solar panels installed on the

roof for local electricity generation.

3.1 Power Consumption

Electrical power is consumed by various loads used

at the household. The house in Figure 1 (a) has many

electrical loads (1 - 6) made up of light loads (lights,

computers, television etc.) and heavy loads (washing

machine, clothes dryer, dish washer etc.) The house

meets its electricity requirements from both the elec-

tricity grid as well as the grid-connected solar system.

We have assumed in this study that gas is used for

other loads for cooking, heating and hot water.

A WEB OF THINGS (WOT) APPROACH TO SMART HOUSEHOLD ENERGY MANAGEMENT FOR

SUSTAINABLE LIVING

(a) (b)

Figure 1: (a): A typical house with grid-connected solar panels and several loads, (b): Energy consumed by a few household

loads in a day as a function of time

The operation of the various loads in a household

depends on the lifestyle of the people in the house,

the weather and the season. Figure 1 (b) shows an

example of the power consumption of the household

and of some loads as a function time of day over a 24-

hour period.The loads in Figure 1 (b) are those which

are mostly ﬁxed in time most days, unlike loads such

as washing machine, dryer or air conditioner which

are ’manageable’ loads that may be operated at any

time of the day within its 24 hour period.

A typical pattern of the electrical power generated

over a 24 hour day by a grid-connected solar panel is

shown in Figure 1 (b). The pattern of solar power may

vary depending upon the weather conditions during a

day and the season in a year. It is assumed in the

ﬁgure that the solar panels and the dc to ac inverter

are rated at 1.2 kW.

Electrical power consumed by the various loads in

the house is supplied from both the utility grid and the

solar panels. This study aims to assist a person of the

household to make decisions on the time of the day at

which such manageable loads are operated for either

reducing the electricity cost for the household, or for

a reduction in the carbon pollution for the household.

3.2 Metering, Cost and Emissions

The cost of electricity to a household is based on the

energy consumed by the household from the utility

grid over a billing period, and so is the greenhouse

emission for the household. An energy meter inte-

grates over time the consumed power as a function

of time to give energy as a function of time (Figure

1). In Victoria, Australia, net metering policy is used

for houses that have grid-connected solar panels for

local electricity generation. Figure 1 shows the net

metering scheme for pricing electricity consumption

by utilities. In Net Metering, both the Grid and the

Solar are taken as sharing the connected loads at any

instant of time. Both the energy ﬂowing into the loads

and the energy ﬂowing into the mains from the solar

system are metered. If the solar power is less than

the connected loads at any time, any additional power

above the solar power is metered as ﬂowing from the

mains. If solar power is greater than connected loads

at any time, additional power from solar is metered as

feeding into the mains exported to the grid.

Power utilities are not necessarily aware of spe-

ciﬁc household energy demands and sustainable liv-

ing objectives in detail of their customers. They in-

stall solar panels and associated infrastructure equip-

ment and wire the system appropriately as per reg-

ulation requirements to enable the power company

to charge the customer for their electricity consump-

tion over the billing period. Currently, the electric-

ity billing system of a power company caters for the

inclusion of the details of the solar energy exported

from the solar panels and the total energy consump-

tion by the entire household for a billing period of two

months. However the bill does not detail a breakdown

of the solar energy production and energy consump-

tion by individual devices in hourly, daily, weekly,

monthly fashion.

Currently, utilities charge a household for the net

energy consumed from the utility grid at 21.6 cents

(Australian) per kWh when the power is consumed

during peak hours of 7:00 am to 11:00 pm and at 10.8

cents per kWh for off-peak consumption. The house-

hold is compensated at 62 cents per kWh for solar

energy exported from the house to the utility. This

rate is used to reduce the customer’s bill amount for

electricity usage. This can be seen as an incentive to

promote sustainable living strategy.

A large proportion of the electricity generated in

PECCS 2012 - International Conference on Pervasive and Embedded Computing and Communication Systems

Victoria is from brown coal, thus resulting in green-

house emissions. So the energy consumed by a house

hold from the utility grid has associated greenhouse

emissions. In Australia, typical values of greenhouse

emission for electricity generation from coal vary be-

tween 0.8 - 1.3 tonnes CO2 equivalent per MWh of

electrical energy generated (Talberg, 2011). We have

assumed a value of 1.2 kg CO2e per kWh for brown

coal.

3.3 Customizing Sustainable Living

The objective of this study is to assist consumers to

actively engage in a software strategy of efﬁcient en-

ergy management by controlling and operating heavy

loads from a dynamic consideration of various fac-

tors, such as solar energy generation, energy costs,

greenhouse emissions, user requirements etc. In order

to create a sustainable household, a customer needs to

measure in real-time the power generated by the solar

system and the power consumed by various loads on

an individual basis as a function of time. The strat-

egy must use the collected real-time data as well as

past data, information on cost and greenhouse emis-

sion and, possibly, weather predictions from the web

to display relevant information to the customer so as

to enable making informed decisions on energy usage.

For example, a customer may require adapting on the

ﬂy running of a heavy load during day time on a sum-

mer day when solar energy production is signiﬁcant.

This may not be feasible because of weather condi-

tions. One may then need to calculate ahead of time

the power used by the chosen machine (say, washing

machine and dryer) and use the information to decide

if an adaptation strategy of using a mix of solar and

grid power for washing, or operate the machine dur-

ing night time when off-peak tariffs apply. One there-

fore needs to measure the power and energy usage as

a function of time and store this information for later

use. Another customer may require information based

on past data to analyze emissions and costs incurred,

and there upon decide a change in lifestyle or use of

energy-efﬁcient technologies and appliances.

4 SMART ENERGY

MANAGEMENT

4.1 System Architecture

The architecture for our case study in smart energy

management for sustainable living is made up of four

layers as shown in Figure 2. The device layer is made

up of physical devices (solar system, appliances etc.)

that we have included in our energy management sce-

nario. The devices are plugged to the Ploggs, which

are smart meter plugs (kWh) and data loggers at the

sensing layer (www.plogginternational.com). In the

Gateway layer, these plogg nodes are discovered and

manipulated by software installed on an embedded

device with Zigbee or Bluetooth communication ca-

pabilities. The gateway includes a micro webserver

and provides RESTful API. The client layer software

enables the householders to visualize their power con-

sumptions in various formats as shown in Figures 2 on

mobile phones or on the web.

4.2 Devices and Sensing Layers

The devices and sensing layers provide an intercon-

nection to other layers by providing information back

and forth on the measured power levels of individual

physical devices as a function of time at prescribed

sampling rates so as to accurately compute energy

as a function of time from power values. The layers

also provide appropriate control information to man-

age the operation of individual devices. The physi-

cal devices consist of power generating devices (so-

lar panels) and loads or appliances in the house hold.

Regulations do not allow customers to install a sen-

sor on the metering system of the utility-grid because

the metering system is an accurately calibrated sys-

tem that is owned by the utility. On the other hand,

the utility may not be able to measure or control indi-

vidual appliances in the household because of privacy

laws.

As shown in Figure 2, each heavy load has a sen-

sor to measure the power consumed by each device as

a function of time. A few light loads may be grouped

to form a cluster that may be connected to one sensor.

A sensor is attached to the inverter connected to the

solar panels to read the power generated as a function

of time by the solar panels. Such a sensing layer en-

ables the discrimination of the power used by individ-

ual devices. As no sensor can be attached to the me-

tering system of the utility, the software system must

be able to compute the total power consumed by the

house hold, the solar power exported to the grid and

the power consumed from the grid at any instant time

from the measured values of power consumed by in-

dividual devices and the power generated by the solar

system.

Monitoring the power consumption

of individual appliances is possible with

Ploggs, a product of Energy Optimisers Ltd

(www.energyoptimizersdirect.co.uk), as shown in

Figure 2. Ploggs may be plugged into electrical

A WEB OF THINGS (WOT) APPROACH TO SMART HOUSEHOLD ENERGY MANAGEMENT FOR

SUSTAINABLE LIVING

Figure 2: Four layer system architecture for energy management.

outlet to measure the power consumption of a

connected appliance in real-time and/or monitor it

over a longer time frame. It can also control the

power consumption remotely over a network. Its

internal clock allows the computing and monitoring

of real-time energy rates as well. Time intervals can

be set from every minute to once a month. Plogg is

based on Zigbee wireless standard, which is used by

various smart meter technologies. Plogg stores the

measured electrical consumption data and wirelessly

transmits the information to a smart phone, computer

or system management software via the internet for

further manipulation.

4.3 Gateway and Applications Layers

In the Gateway layer, Plogg nodes are discovered and

manipulated by the software installed on a device with

Zigbee or Bluetooth communication capabilities as

shown in Figure 2. The Gateway software discovers

the Ploggs by scanning the environment and makes

them available as web resources. The gateway in-

cludes a micro webserver, which enables access to

Ploggs over the web, and allows the management of

Ploggs as structured URLS in a RESTful style.

The Gateway software communicates with the

Ploggs and delivers to Applications layer the results,

either as JSON (JavaScript Object Notation) docu-

ments or as HTML representation of power and en-

ergy consumption of devices. A RESTful web API is

developed as part of the Gateway layer. A RESTful

web API is a web service implemented using HTTP

and four REST principles of addressability, state-

lessness, connectedness and uniformity (Erenkrantz,

2009; Caporuscio et al., 2011). It is hypertext driven

and supports operations using HTTP methods (GET,

PUT, POST, DELETE) and supports media types

such as JSON (JavaScript Object Notation), XML

etc. REST has an important concept called Resources,

each of which is referenced with a global identiﬁer

(e.g. URI in HTTP).

The Application layer shown in the architecture

(Figure 2) enables the user to visualize information

in various formats on mobile phones or on the web.

Modelling for context and for activity are combined

as user preference in the user interface (Romero et al.,

2010). The contexts under consideration in the energy

management system are: weather, month, time of day,

location (at home, away from home). At the request

of the user, the application may display the power

consumed by individual loads or the solar power gen-

erated over a day. It should also be able to display

historic information of power or energy over a cho-

sen period in the past, say over a month 2 years ago.

Location-related remote operations when away from

home are related to delay in coming home and may

involve lights or air-conditioning being switching on

remotely. The activities considered are: laundry,

PECCS 2012 - International Conference on Pervasive and Embedded Computing and Communication Systems

(a) (b)

Figure 3: (a) - Computed results for the total power consumed and solar power. Case 1 - Use of washing machine and dryer

during daytime; (b) - Comparison of results of costs and emissions for the two cases of Washing Machine used.

reverse-cycle air-conditioning, dish washer, small ap-

pliances such as iron, lighting, computer, telephone,

vacuum cleaner, cooker etc.

An example of the power consumption and solar

power generation over a day is shown in Figure 3(a) in

which the heavy load of washing machine and dryer

is operated during the day time when the solar power

generation is high.

The lower ﬁgure in Figure 3(a) shows the com-

puted difference between the total power consump-

tion and solar generation to illustrate clearly the ex-

port of solar power to the grid. In order to make a de-

cision on the time of a similar day in the future when

the washing machine needs to be turned on, the appli-

cation may be asked to display electrical powers over

a day. A comparison of the day time and night time

scenarios considered may also need to be provided by

the applications. Such a comparison is shown in Fig-

ure 3(b).

Based on the results shown in Figure 3, the cus-

tomer may choose an option of reducing cost or

greenhouse emission.

5 CONCLUSIONS

We have presented in this paper a four-layered archi-

tecture for monitoring and control of web-enabled de-

vices. It uses the Internet of Things approach where

the focus is on interaction and interoperation between

smart devices and between devices and people. This

has been achieved through 6LoWPAN standard that

includes IP for smart devices and enables the efﬁ-

cient use of IPv6 over low power low rate wireless

networks. Physical devices use the sensing layer to

communicate with the Ploggs, which can monitor and

control the devices. The devices and sensing layers

form the lower two levels of the architecture. The

key aspects of Web of Things are to integrate smart

devices into the web and abstract integrated smart

devices to web services. RESTful web service has

been presented as part of the architecture layer so-

lution. The Gateway layer and the application layer

interaction enable the information to be presented to

suit varying the consumer requirements. For exam-

ple, some consumers may want real-time information

whereas others may want to look at past archived data

to make decisions.

A case study of electricity management has been

considered in this paper. It allows the consumer to

use the past and predictive information of appliances

and manage the operations of appliances efﬁciently

to achieve cost reduction or GHG emission reduction.

The results of electricity costs and GHG emissions

are presented for operating a heavy load (washing ma-

chine and dryer) in a house in Melbourne, Victoria at

two different times of a day when solar energy pro-

duction is signiﬁcant. The presentation of such re-

sults can assist the consumers in making decisions

to change their lifestyle in using electrical devices at

times of the day when cost or GHG emissions reduc-

tion can be achieved. We believe that the architecture

presented can be extended to include other energy re-

sources, such as gas, wind and stored solar energy. It

can also be extended to include water management at

household level where houses receive water from the

mains as well as from stored water from rain water

tanks.

ACKNOWLEDGEMENTS

The work reported in this paper was conducted par-

tially at the Dependable Evolvable Pervasive Soft-

A WEB OF THINGS (WOT) APPROACH TO SMART HOUSEHOLD ENERGY MANAGEMENT FOR

SUSTAINABLE LIVING

ware Engineering Group, Politecnico di Milano by

the lead author during her research fellowship from

May to July 2011. The author thanks Prof Carlo

Ghezzi and Dr Mauro Caporuscio for the discussions

held during her fellowship visit.

REFERENCES

Abdelzaher, T. (2006). Towards an architecture for

distributed cyber-physical systems. In Proceed-

ings of NSF Workshop on Cyber-Physical Systems.

http://varma.ece.cmu.edu/Summit/CPS.

Caporuscio, M., Funaro, M., and Ghezzi, C. (2011). REST-

ful service architectures for pervasive distributed en-

vironments, chapter REST:From Research to Practice.

Springer.

Design, E. T. (2011). The wireless embedded internet,

part i. http://www.eetimes.com/design/embedded-

internet-design/4216240/6LoWPAN.

Dillon, T. S., Zhuge, H., Wu, C., Singh, J., and Chang, E.

(2011). Web-of-things framework for cyber-physical

systems. Concurrency and Computation: Practice

and Experience, 23:905–923.

Erenkrantz, J. R. (2009). Computational REST: A New

Model for Decentralized, Internet-Scale Applications.

PhD thesis, University of California, Irvine.

Guinard, D. (2010). Mashing up your web-enabled home.

In Adj. Proc. The International Conference on Web

Engineering (ICWE 2010), Vienna, Austria.

Guinard, D., Trifa, V., and Wilde, E. (2010). A resource ori-

ented architecture for the web of things. In Proceed-

ings of IEEE Conference on the Internet of Things.

Hui, J. W. and Culler, D. E. (2008). Ip is dead, long live ip

for wireless sensor networks. In SenSys ’08: Proceed-

ings of the 6th ACM Conference on Embedded Net-

work Sensor Systems, pages 15–28, New York.

Kamilaris, A., Trifa, V., and Pitsilledes, A. (2011). The

smart home meets the web of things. Int. Journal of

Adhoc and Ubiquitous computing, (IJHUC), Special

issue on the smart digital home, 7(3):145–154.

Ostermaier, B., Schlup, F., and Romer, K. (2010). Webplug:

A framework for web of things. In Proceedings of

PerCom Workshops 2010, pages 690–695.

Papadimitriou, D. (2009). Future internet - the cross-etp vi-

sion document. http://www.future-internet.eu. version

1.0.

Richardson, L. and Ruby, S. (2007). Restful Web Services.

O’Reilly Media, 1st ed. edition.

Romero, D., Rouvoy, R., Seinturier, L., and Loiret, F.

(2010). Integration of heterogeneous context re-

sources in ubiquitous environments. In SEAA’10.

Shelby, Z. and Bormann, C. (2010). 6LoWPAN: The wire-

less embedded Internet. Wiley publishing.

Talberg, A. (2011). Performance standards to reduce en-

ergy emissions. Parlimentary Library, Parliament of

Australia.

Zeng, D., Guo, S., and Cheng, Z. (2011). The web of

things (an invited paper). Journal of Communications,

6(6):424–43.

PECCS 2012 - International Conference on Pervasive and Embedded Computing and Communication Systems