Taxonomy for Internet of Things

Tools for Monitoring Personal Effects

Leon Barker, Martin White, Mairead Curran, Zeeshan Patoli, Benji Huggins,

Tudor Pascu and Natalia Beloff

University of Sussex, Department of Informatics, Chichester 1, Brighton, U.K.

Keywords:

Internet of Things (IoT), User Experience Design, Ubiquitous Computing, Big Data, Peer-to-Peer Networks,

Cloud Computing.

Abstract:

Human capacities to track and identify peers, products and activities have beneﬁted from the wide proliferation

of networked electronic devices; developments in smart-materials, near-ﬁeld communication and computer-

vision technology further advance these capabilities. During this investigation into The Internet of Things

(IoT) a new taxonomy was created together with a set of prototype applications and accompanying archi-

tecture. In addition to presenting these taxonomies the creation of a peer-to-peer network with a distributed

database is discussed. The development of a system that enables users to track and exchange objects or services

using a secure and robust data repository model is also proposed.

1 INTRODUCTION

The research presented in this paper initially focused

on investigating the types of activities, applications

and behaviours that may result from the emergence of

ubiquitous smart devices and the establishment of the

Internet of things (IoT) infrastructure. Prototype ap-

plications have been created as a result and user trials

are being conducted. This research was undertaken

as part of a UK interdisciplinary Technology Strategy

Board (TSB) Collaborating Across Digital Industries

2 (CADI2) project ’Connecting virtual communities

to the digital economy through micropayment tech-

nologies’.

The IoT concept is not new; though originally pro-

posed in 1999 (Ashton, 2009) and often discussed

within the ﬁelds of Human Computer Interaction

(HCI) there is little consensus regarding the compo-

sition and application of this concept. The lack of

clear consensus creates obstacles that prevent holis-

tic perspectives on the impact and composition of IoT

infrastructures from being gained. This lack of con-

sensus serves only to further reduce the capacity for

interdisciplinary ﬁelds of research and industry to col-

laborate in the development of products and the evalu-

ation of IoT systems. The outcome of such consensus

would aid the development of a robust IoT system ar-

chitecture. The creation of such a framework would

provide a useful template to support and inspire de-

velopers to create novel IoT applications that are of

genuine beneﬁt to users.

The wide distribution of smart-devices not only

allows users access to information regarding cur-

rent and historic global events they also facilitate

users with the ability to communicate with their

peers. Moreover, through advances in technology

users will soon be able to monitor there own bio-

signals through the use of smart-fabrics such as

temperature-responsive polymers (Crespy and Rossi,

2007), shape-memory polymers (Nji and Li, 2010)

and polyelectrolytes coated with carbon nanotubes

(Shim et al., 2008).

The rapid development of smart objects that have

the capacity to access and broadcast sensitive infor-

mation make the establishment of a consensus regard-

ing the direction of IoT both signiﬁcant and important

to address. The establishment of a coherent IoT archi-

tecture would provide positive beneﬁts to end-users of

smart technologies by enabling the creation of agreed

protocols that will help provide consistent user expe-

riences and a layer of data security.

Though there are many obvious beneﬁts to es-

tablishing a robust IoT system architecture there is

also signiﬁcant potential for conﬂict to occur between

competing forces within technology and social sec-

tors. As IoT systems by their nature are dynamic they

can be conﬁgured in multiple ways and can either be

networked as a centralised data repository or a decen-

Barker L., White M., Curran M., Patoli Z., Huggins B., Pascu T. and Beloff N..

Taxonomy for Internet of Things - Tools for Monitoring Personal Effects.

DOI: 10.5220/0004809500670071

In Proceedings of the 4th International Conference on Pervasive and Embedded Computing and Communication Systems (PECCS-2014), pages 67-71

ISBN: 978-989-758-000-0

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

tralised peer-to-peer network. The potential ramiﬁca-

tions of each system conﬁguration is outlined in sec-

tion 2.2 of this paper.

Through the course of investigating the current

IoT landscape a wide range of interconnected ele-

ments and components were identiﬁed. In an effort to

provide a coherent frame of reference a set of taxon-

omy for describing the active elements observed dur-

ing this investigation have been deﬁned. This taxon-

omy is described in section 2.1 .

2 TAXONOMY OF THINGS

It is common for discussions regarding IoT to focus

upon only one category of object; it is generally as-

sumed that an IoT eco system is made up of net-

worked electronic devices. Advances in near-ﬁeld

communication and computer-vision equip modern

day smart phones with the ability to track and iden-

tify a wide variety of objects. This capability means

a wide array of different object types can be contain

within the IoT ecosystem.

2.1 IOT COMPONENTS

The categories of objects constituted within an IoT

ecosystem have been observed as follows:

• Smart Devices;

• Persistent Nodes;

• Collectables;

• Actor/Sentient Agent;

• Semi Autonomous Agent;

• Loose Perishables;

• Gateways.

2.1.1 Smart Devices

The smart device in our IoT ecosystem describes a

device or mechanism that is able to communicate its

own presence or existence to servers and other lis-

tening devices over either a wired or wireless com-

munication network. Typically a smart device also is

equipped with sensors and is able to broadcast infor-

mation about its local environment. Smart devices are

generally assigned a unique ID, which they are able to

both store and communicate.

2.1.2 Persistent Nodes

The Persistent node can exist in the mid to long term.

It is part of a network and typically provides a service

or storage function. These nodes have unique IDs,

buildings and sensors can fall into this category. A

warehouse can be a persistent node in a supply chain

just as a weather vane can be a persistent node in a

meteorological survey. Persistent nodes tend to op-

erate in closed networks that access the outer world

through gateways.

2.1.3 Collectables

Collectables can represent a wide array of objects

such as books, teacups cornﬂake packets and any ob-

ject that have been assigned a UPC barcode. These

objects do not have unique IDs because it is imprac-

tical, as it would only succeed in creating a bloated

database and no real performance beneﬁts. It is only

through association with other IoT object types that

collectables can become unique.

2.1.4 Actors/Sentient Agent

Actors and Sentient Agents describe things that are

by nature unique and are capable of acquiring col-

lectable things. Sentient things exhibit free will and

thus are able to make unprompted decisions about

their actions. People obviously fall into this category

and perhaps in the future some machines will also be

considered sentient.

2.1.5 Semi Autonomous Agent

A semi autonomous agent is only capable of initiating

actions once a super-user or systems admin has set

speciﬁc permissions for a particular action. Once an

action has been authorised a semi autonomous agent

is capable of acquiring collectables and conducting

tasks in a similar way to sentient agents.

2.1.6 Loose Perishable

Loose perishables consist of objects such as fruit and

unpackaged foods stuff; it also describes objects with

a limited lifecycle that would be impractical to assign

an individual unique ID. Asides from their limited

lifecycle loose perishables have the same character-

istics as collectables.

2.1.7 Gateways

Gateways are essential parts of any IoT ecosystem as

they are the link between interconnected networks. A

router is the gateway that allows computers to access

the Internet, just as the point-of-sale device (POS)

is the gateway that allows banks and individuals to

transfer funds to retailers and other such organisa-

tions. Gateways are the links between nodes and are

PECCS2014-InternationalConferenceonPervasiveandEmbeddedComputingandCommunicationSystems

Figure 1: Category-One decentralised peer-to-peer net-

work.

Figure 2: Category-Two centralised gateway repository.

therefore the key elements to consider during the de-

velopment of security and privacy protocols.

It is possible for particular objects to ﬁt more than

one of the categories described, for example a loose

perishable object could also be collectable for the du-

ration of its lifespan. However once a thing is cate-

gorised it becomes easier to see how objects function

within an IoT ecosystem.

2.2 Central and Decentralised Networks

Different network conﬁgurations are possible within

an IoT ecosystem; for example as an IoT ecosystem is

composed from many components it is dynamic and

can be conﬁgured as either a centralised data repos-

itory or a decentralised peer-to-peer network. Cur-

rently, companies seeking to proﬁt from IoT are at-

tracted towards centralised cloud based architecture

where data is stored in a central repository. This ap-

proach offers great potential and opportunity for com-

panies to capture large amounts of data regarding the

activities and habits of consumers. The emergence

of big-data analytic tools currently demonstrates that

messy data can be analysed to facilitate insight re-

garding the habits of network users. This type of ar-

chitecture reﬂects a category-two (see ﬁgure: 2) in-

frastructure (Karimi and Atkinson, 2013), where all

end nodes send data to centralised repositories. This

type of network can draw upon pre-existing IT infras-

tructure where servers are centrally located in dedi-

cated server farms throughout the world. This type of

approach seems to ignore some of the novel attribute

offered by an IoT ecosystem. One of the main at-

tributes of IoT is the notion of the hyper-connectivity

of devices. We are currently seeing the wide distribu-

tion of devices capable of communicating with other

systems both globally and locally. The centrally lo-

cated cloud based server that is seemingly being ad-

vocated by meanly companies ignores the potential

of a local peer-to-peer IoT infrastructure. This type of

architecture reﬂects a category-one infrastructure (see

ﬁgure: 1) (Karimi and Atkinson, 2013), which en-

compasses a heterogeneous architectural model con-

sisting of millions of peers.

By insisting that a device report back to a server

that is often hundreds if not thousands of miles

away we have ladened our personal devices with en-

ergy hungry telecommunications chips, which require

these devices to be continually recharged. A lot of en-

ergy could be saved if our devices only had to interact

with local networks. Through a single search query it

is claimed that on average Google reads one hundred

megabytes of data, which consumes tens of billions

of CPU cycles (Barroso et al., 2003). Added to these

processing requirements is the need to send the data

via satellite and wired connections as well as server

cooling issues; these factors illustrate the impact that

a centralised repository has on power consumption.

To address the need for the creation of a peer-to-

peer network that allows for the secure, accurate and

robust transfer of sensitive and private information a

modular data storage infrastructure needs to be cre-

ated. A prototype infrastructure was instigated dur-

ing the course of this research investigation. The pro-

posed infrastructure will enable the transfer of infor-

mation in local or global peer-to-peer exchange net-

works through widely distributed repositories.

TaxonomyforInternetofThings-ToolsforMonitoringPersonalEffects

3 IOT APPLICATION

PROTOTYPE

There currently is no decentralised mechanism that

enables mobile phone users to securely acknowledge

the receipt and exchange of items. There are also few

applications to facilitate the seamless transfer of lists

between devices. Should a user wish to compile a lists

of items such as ingredients for a recipe or electronic

components to build a circuit, the user has to labori-

ously search for each individual item via third-party

websites and inventories. The application developed

during this project provides a mechanism for pushing

lists to peers and merchants, whilst also providing a

secure and seamless peer-to-peer record of exchange.

The emergence of hyper-connected objects with

their own digital identities within IoT landscapes

presents new opportunities for users to track and

access information regarding their own possessions.

The mobile application and supporting infrastructure

presented has been designed to allow users to eas-

ily compile and access lists of owned objects, retail

products and other things. Users are able to capture

items via their smart phones using NFC, QR codes,

barcodes, image-recognition or by manually entering

the items description. Once a list has been compiled

the user is able to push the list through desired chan-

nels. Using the application users can generate lists of

recipe items and then purchase the item through an

in-app shopping channel, by using the API of par-

ticipating partners (see ﬁgure: 3). Alternatively a

user might compile a list of items they plan to sell

and use the app to list their items in auction or recy-

cling channels such as Ebay or Freecycle. Using the

framework developed users can even use the applica-

tion to share books, music playlist, games and DVDs

amongst friends, family and the wider community. In

future scenarios users might store preferences for how

their smart-fabrics and smart-vehicles should respond

under speciﬁc social or climate conditions.

During the development of this demo applica-

tion a peer-to-peer object archive framework was de-

veloped. The resulting application enabled users to

seamlessly monitor possessions through their entire

lifecycle (from purchase to disposal). Though ini-

tial development activity focussed on the redesign of

a physical shopping basket it later became clear that

the creation of a receipt repository for tracking gen-

eral peer-to-peer transaction and interaction could of-

fer wider beneﬁts. Though an ever-increasing number

of monetary transactions are conducted electronically

there are few applications to facilitate seamless peer-

to-peer transactions. The applications that do exist

do not provide users with a method of recording the

type of exchange both parties enter. The prototype ap-

plication created through this research enables users

to keep a record of past transactions through a dis-

tributed data store.

Figure 3: Category-Two centralised gateway repository.

4 CONCLUSIONS

The creation of the demo application is a trivial soft-

ware development task, however the creation of a se-

cure and robust decentralised repository that is dis-

tributed among a large peer-to-peer network is a sig-

niﬁcant challenge. We are beginning to see the emer-

gence of exchange systems that exhibit the necessary

characteristics to support decentralised peer-to-peer

exchange. For example Bitmessage (Warren, 2012),

Bitcoin (Nakamoto, 2008), Litecoin (Sprankel, 2013)

and Namecoin represent cryptographic currencies and

communication tools that offer decentralised peer-to-

peer transactions. However these mechanisms do not

include information regarding the item that has been

exchanged and are designed to offer anonymity to

both seller and buyers. The model we are develop-

ing would record the type of agreement each party

has entered into as well as providing both parties with

a formalised proof of exchange.

ACKNOWLEDGEMENTS

This research was undertaken as part of a UK interdis-

ciplinary Technology Strategy Board (TSB) Collab-

orating Across Digital Industries 2 (CADI2) project

’Connecting virtual communities to the digital econ-

omy through micropayment technologies’.

REFERENCES

Ashton, K. (2009). That internet of things thing in

the real world, things matter more than ideas.

RFID Journal. Retrieved from http://www. rﬁdjournal.

com/article/articleview/4986/1/82.

PECCS2014-InternationalConferenceonPervasiveandEmbeddedComputingandCommunicationSystems

Barroso, L. A., Dean, J., and Holzle, U. (2003). Web search

for a planet: The google cluster architecture. Micro,

Ieee, 23(2):22–28.

Crespy, D. and Rossi, R. M. (2007). Temperature-

responsive polymers with lcst in the physiological

range and their applications in textiles. Polymer In-

ternational, 56(12):1461–1468.

Karimi, K. and Atkinson, G. (2013). What the Internet of

Things (IoT) Needs to Become a Reality. . Technical

report, Freescale.

Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic

cash system. Consulted, 1:2012.

Nji, J. and Li, G. (2010). A self-healing 3d woven fab-

ric reinforced shape memory polymer composite for

impact mitigation. Smart Materials and Structures,

19(3):035007.

Shim, B. S., Chen, W., Doty, C., Xu, C., and Kotov, N. A.

(2008). Smart electronic yarns and wearable fab-

rics for human biomonitoring made by carbon nan-

otube coating with polyelectrolytes. Nano letters,

8(12):4151–4157.

Sprankel, S. (2013). Technical basis of digital curren-

cies. Retrieved from http://www.coderblog.de/wp-

content/uploads/technical-basis-of-digital-

currencies.pdf.

Warren, J. (2012). Bitmessage: A peer to peer message

authentication and delivery system. Retrieved from

https://bitmessage.org/bitmessage.pdf.

TaxonomyforInternetofThings-ToolsforMonitoringPersonalEffects