Internet of Bicycles
Tracking and Monitoring Life-cycle Information using GS1
Miyeon Lee
1, 2
, Sunghoon Lee
2, 3
, Jaehyung Choi
2, 4
, Seongsik Kim
2
and Daeyoung Kim
2
1
LG Electronics, Seoul, Republic of Korea
2
Software Graduate Program, KAIST, Seoul, Republic of Korea
3
KEPCO KDN, Naju, Republic of Korea
4
Samsung, Suwon, Republic of Korea
Keywords: Internet of Things, IoT, Bicycle, Global Standard 1, GS1.
Abstract: Social phenomena such as the rise of cyclists, the expansion of public bicycle systems and the increase of
bicycle thefts highlight the needs of tracking a bicycle’s life-cycle and implementing new services based on
information from a bicycle’s life-cycle. We suggest Global Bicycle Information Architecture to describe
and save a bicycle’s life-cycle. We extend the GS1 EPCglobal architecture for Global Bicycle Information
Architecture to identify bicycles and capture and share information. Global Bicycle Information
Architecture enables stakeholders to gather information that can be used to formulate public policies or for
protection of property – in this study, bicycles. We verify the availability of Global Bicycle Information
Architecture with the implementation of a bicycle tracking system.
1 INTRODUCTION
Even though there is no generally accepted
definition of the Internet of Things (IoT), IoT
generally refers to a network of physical or virtual
things that enables these objects to collect and
exchange data. The network connectivity of “things”
means that IoT finds applications almost every
industry including healthcare, utilities and transport
(Sundmaeker et al., 2010). According to Gartner,
Inc., there will be nearly 26 billion devices on IoT
by 2020 (Gartner, 2013).
Intelligent bicycles become a part of smart cities
since they exchange information with traffic systems
(e.g. traffic lights, public transportation system) and
connected cars to improve the citizen life, help
avoiding/reducing accidents and so on (Dezani,
2015).
In addition, many countries have public bicycle
systems, known as bicycle-sharing systems. From
2000 to 2014, the number of cities with a bicycle-
sharing system increased by approximately 214
times (Russell, 2015). However, these systems
cannot be connected with each other because they
use different identification systems for the users and
bicycles (Erlanger, 2009).
However, bicycle thefts also have increased.
According to statistics from the United Kingdom,
536,166 cases were reported from 2008 to 2013
(Moss, 2014). But only 1 in 4 bike thefts are
reported to the police. This means that almost 2.1
million bicycles are stolen over five years. These
social phenomena show the need for tracking a
bicycle’s life-cycle and implementing new services
based on information from the life-cycle.
In 2014, the Netherlands Organisation for
Applied Scientific Research released a prototype of
the country's first intelligent electric bicycle, which
could be available to consumers within the next two
years according to The Telegraph (Telegraph Men,
2014). Similar to recently released connected cars,
the bicycle is designed to warn its rider of oncoming
dangers with electronic devices. This approach is
aimed at reducing accidents, but does not try to
connect other bicycles or share information.
Recently connected bicycles also have been
studied. Connected bicycles exchange information
with other connected bicycles to help cyclists
avoiding traffic jams or maintaining a safe distance.
But this approach only focuses on the connection
between bicycles (Cespedes et al., 2014).
The present research suggests Global Bicycle
Information Architecture which is an Internet of
Lee, M., Lee, S., Choi, J., Kim, S. and Kim, D.
Internet of Bicycles - Tracking and Monitoring Life-cycle Information using GS1.
DOI: 10.5220/0005876500570064
In Proceedings of the International Conference on Internet of Things and Big Data (IoTBD 2016), pages 57-64
ISBN: 978-989-758-183-0
Copyright
c
 2016 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
57

Things framework for bicycles as a solution of these
social phenomena. This framework provides a model
of activities in a bicycle’s life-cycle and enables
stakeholders to capture and share information about
the life-cycle of bicycles based on Global Standard 1
(GS1).
For decades, GS1 has provided multiple
standards that enable industries to identify items and
capture and share information from their identifiers.
GS1 suggests three steps based on the life-cycle
of information – Identify, Capture and Share. These
three steps enable information to be uniquely
identified and captured with a general format and
shared automatically. This concept is also applied to
the Internet of Bicycles to identify bicycles, capture
event data during the life-cycle of bicycles and share
data with other users.
Global Bicycle Information Architecture also
enables bicycles to connect with other bicycles,
mobile phones, cars and a traffic control system to
give people information and protect people’s
property – bicycles in this research. This information
can be used for a country’s bicycle-registration
policy, a recall service and other security services to
track bicycles with a globally unique identifier.
This paper is organised as follows: Section 2
introduces the GS1 architecture, which gives the
basic structure of Global Bicycle Information
Architecture. Section 3 provides Global Bicycle
Information Architecture suggested in this study. In
section 4, we show the practical use of Global
Bicycle Information Architecture by implementing a
bicycle tracking service and finally conclude our
paper in section 5.
2 GLOBAL BICYCLE
INFORMATION
ARCHITECTURE
This research provides an Internet of Bicycles
named Global Bicycle Information Architecture.
Global Bicycle Information Architecture is an
architecture that enables stakeholders to identify
bicycles and capture their information and share
information to other stakeholders.
Information from bicycles is usable for
manufacturers to manage their products, retailers to
manage their goods, end users to track their
properties and governments to manage a public
bicycle sharing system and order recall. Also this
information can be an opportunity to service
providers.
In this architecture, we generalize a bicycle’s
life-cycle in four steps – manufacture, sell,
public/private use and disuse. Each step has one or
more stakeholders who generate a business step and
make information or events. This architecture
consists of stakeholders and their systems to capture
information and Electronic Product Code
Information Service (EPCIS) elements to share
information.
In section 2.1 and 2.2, this paper introduces
Electronic Product Code (EPC) Network
Architecture and Oliot which are based on Global
Bicycle Information Architecture. Section 2.3
provides Global Bicycle Information Architecture
including the information needs based on a bicycle’s
life-cycle. This paper describes a bicycle’s life-cycle
in four steps and suggests an abstract model of each
step in section 2.4. This model enables stakeholders
in the architecture to share and understand
information equally.
2.1 EPC Network Architecture
The EPC Network Architecture provides the basic
structure for the Internet of Bicycles. There are two
major elements that are different from a standalone
database system:
 EPCIS
 Object Naming Service (ONS)
To gather information, people need to identify a
target object. For this reason, GS1 provides eleven
categories to give globally unique keys such as
trading, location or shipping.
For the Internet of Bicycles, this research uses
Global Trade Item Number (GTIN) because the life-
cycle of bicycles starts from the manufacture of the
bicycle and GTIN gives an advantage to describe
products as US food supply has shown (Krissoff et
al., 2004).
After identifying a target thing, EPCIS can
capture information by its capturing applications and
share information by its access applications. To
query information to other EPCIS all over the world,
EPC Network Architecture provides ONS to find
where queried information is.
The Internet of Bicycles described in this paper
also uses the EPC Network Architecture to share
information from bicycles such as their status (e.g.
in stock, parked, being ridden or stolen), whether
they are registered as public bicycles or if they have
been stolen.
IoTBD 2016 - International Conference on Internet of Things and Big Data
58

2.2 Oliot: Open Language IoT
Oliot is an open source project to build an ID-centric
IoT Platform developed at Auto-ID Lab in KAIST.
Oliot implements the platform based on GS1
identification standards such as URI-convertible
GTIN.
Oliot enables identification of intelligent things
and capturing and sharing of information with each
other. Oliot supports passive and active tags, various
sensors and actuator networks (e.g. Barcode, Zigbee,
6LoWPAN). Oliot has an approach that integrates
IoT-related projects. For instance, Oliot EPCIS
shows possibility of application to healthcare (Byun
and Kim, 2015).
The Internet of Bicycles described in this paper
is derived from Oliot’s concept.
2.3 Global Bicycle Information
Architecture
To identify, capture and share information about
bicycles, Global Bicycle Information Architecture
consists of three major actions, as the GS1
architecture describes:
 ‘Identify’ bicycles
 ‘Capture’ information from bicycles
 ‘Share’ information to others
This architecture uses an identification policy based
on SGTIN. GTIN is an identifier for trade items and
it enables manufacturers to describe a product easily.
GTIN includes a company and a product reference
number. GTIN is also helpful to order a recall to
manufacturers and retailers because a product
reference number in GTIN generally includes a
model number. To identify a specific bicycle, this
architecture uses SGTIN to identify it and provides a
manufacturer’s original serial number as an
extension.
Information that Global Bicycle Information
Architecture captures should be useful for general
purposes. For example, manufacture and sell
information is necessary when a government orders
a recall to manufacturers and retailers. Riding or
parking (use) information is useful when a
government appropriates budget to build bicycle
roads or parking facilities.
To capture and share information about bicycles,
this architecture includes EPCIS and ONS. EPCIS
and ONS enable Global Bicycle Information
Architecture to generate and save information based
on the global standards.
Figure 1: Global Bicycle Information Architecture.
Figure 1 shows an overview of Global Bicycle
Information Architecture. The stakeholders generate
information from four business steps – manufacture,
sell, public and private use and disuse. The
generated information is captured by EPCIS
capturing applications installed in the stakeholders’
environment. The captured information is stored in
EPCIS. ONS guides applications in terms of where
they need to save information and where they can
find other information.
2.4 Bicycle Object and Process
Modelling
Bicycles are used as ‘objects’. This means that a
bicycle and its related events should be generalized
and abstracted. In this section, we describe necessary
steps and define what should be saved. For example,
if a bicycle is registered as a public bicycle, the
public enrolment step should be modelled. To
describe this step, the following information is
registered: which bicycle it is, when it is registered,
how long it serves as a public bicycle and where it is
located.
For event data and master data, GS1 Core
Business Vocabulary Standard (CBV) is applied in
this architecture. This architecture also defines new
user vocabularies to describe a bicycle’s events.
2.4.1 Bicycle Process Modelling
For Global Bicycle Information Architecture, we
define four stakeholders: manufacturers, retailers,
governments, and end users. The following eleven
actions are considered as steps of a bicycle’s life-
cycle:
 A manufacturer makes a bicycle.
 A manufacturer sells a bicycle to a retailer.
Internet of Bicycles - Tracking and Monitoring Life-cycle Information using GS1
59

 A retailer sells a bicycle to an end user or a
government.
 A government registers a bicycle as a public
bicycle system.
 An end user registers a bicycle into a
government.
 An end user rents a public bicycle.
 An end user returns a public bicycle.
 An end user rides a (public/private) bicycle.
 An end user stops riding a (public/private)
bicycle.
 A (public/private) bicycle is stolen.
 A bicycle theft is reported to a government
(police office).
 A bicycle ends its life-cycle.
Figures 2 shows a bicycle’s general life-cycle from
production to disuse.
Figure 2: A Bicycle’s Life-cycle and Events Overview.
These actions become event data that are saved
in EPCIS:
Table 1: Event and its description.
Event Description
Production A manufacturer makes a bicycle.
Selling
A retailer sells a bicycle to a
government or an end user.
Public enrolment
A government enrols a bicycle for
public purposes.
Private enrolment
An end user enrols a bicycle to a
government.
Disuse A bicycle ends its life-cycle.
Robbery reporting
A bicycle theft is reported to a
government (police office).
Robbery A bicycle is stolen.
Riding
An end user rides a bicycle. A public
bicycle is rented when an end user rides.
Parking
An end user stops riding a bicycle. A
public bicycle is returned when an end
user stops riding.
Recall
A government orders a recall campaign
to a manufacturer or a retailer.
Each event is described in the next section.
These actions can be extended, especially 3
rd
party
services such as insurance, a security service, and so
on. Furthermore, each event is translated as a
business step keyword described in Table 2.
Table 2: Event and its business step.
Event Business Step Subject
Production manufacturing Manufacturer
Selling retail_selling* Retailer
Public
enrolment
public_enrol
(add/delete)
Government
Private
enrolment
private_enrol
(add/delete)
End user
Disuse destroying*
Government/
end user
Robbery
reporting
robbery_reporting
(add/delete)
Government
Robbery robbery End user
Riding riding End user
Parking parking End user
Recall order_recall Government
(*: defined core business step in CBV)
2.4.2 Bicycle Data/Object Modelling
All information generated by bicycles must be
changed to the format of EPCIS event data to save it
in the EPCIS server. Based on the GS1 architecture,
event data should have the following ‘4W’ structure:
 What
 When
 Where
 Why
In this section, we define eight business steps of a
bicycle (two other business steps are already defined
in CBV), two dispositions and extensions to give
more information. We also suggest a new
namespace ‘bicycle’ to mark these events. These
vocabularies and namespace should be included in
the standard.
2.4.3 4W: What
The “WHAT” dimension indicates the objects to
which the EPCIS event pertains.
Each observed bicycle should be captured in a
separate ObjectEvent. The epcList element should
contain only the SGTIN of the observed bicycle. To
specify a bicycle for its manufacturer, this
architecture also suggests the extension
‘serialNumber’ described in section 2.4.2.4.
2.4.4 4W: When
The “WHEN” dimension serves as a timestamp for
the EPCIS event.
The eventTime of Object events should reflect
the time at which the bicycle was observed. In
IoTBD 2016 - International Conference on Internet of Things and Big Data
60

addition, the recordTime reflects the date and time at
which this event was recorded by an EPCIS
Repository.
2.4.5 4W: Where
The “WHERE” dimension component indicates the
location at which the EPCIS event was observed, as
well as the whereabouts of the object subsequent to
the event.
The readPoint – indicates the SGLN
corresponding to the event’s location. In this
process, it could be a factory, a shop, a government
or a parking facility.
The bizLocation – indicates the SGLN
corresponding to another related location. For
example, a ‘Robbery reporting’ event has a
readPoint of government (police office) and a
bizLocation of the place where the bicycle went
missing.
This architecture also allows an extension that
enables GPS location directly. For all object events
and transaction events either the readPoint or the
bizLocation or both should be specified.
2.4.6 4W: Why
The “WHY” dimension reflects the business context
(“Business Step”) of the EPCIS event, as well as the
status (“Disposition”) of the object subsequent to the
event. The Core Business Vocabulary (CBV) defines
standard values for these.
A Business Step specifies the business process
linked to the EPCIS event. Two of Business Step
identifiers are specified in section 7.1 of the CBV
(GS1, 2014). Other business steps are newly defined
in Table 2.
The Disposition denotes the status of a bicycle,
subsequent to the EPCIS event. The Disposition is
assumed to hold true until a subsequent event
indicates a change of disposition. Disposition
identifiers are specified in section 7.2 of the CBV.
The following disposition identifiers are applied in
this architecture:
Table 3: Disposition and its description.
disposition value Description
enabled
The bicycle’s service or registration
activated
disabled
The bicycle’s service or registration
deactivated
All of these elements should be specified using
the suggested namespace ‘bicycle’ for various
extension elements. These extension elements give
extra information of each event. For example, when
a user registers his bicycle to a government, a user’s
identification is necessary.
Table 4: Extension and its description.
extension value Description
serviceLife
It indicates the durability of the
public bicycle.
serialNumber
It indicates the serial number of the
bicycle generated by a manufacturer.
userIdentifier
It indicates a personal identifier for
the specification of the owner. It
should follow their country’s law.
recall_parts
It indicates a part or a set of parts
related to a recall order. CPID is used
for it.
distance
It indicates the distance between a
bicycle and a rider.
2.4.7 Biportal XSD Example
Figure 3 shows an example of XML Schema
Definition for events in Global Bicycle Information
Architecture. It describes two additional dispositions
and five extensions and these data types (e.g. string,
integer).
Figure 3: XML Schema Definition Example.
3 IMPLEMENTATION
We provide BiPortal service to implement Global
Bicycle Information Architecture and its utilization.
Figure 4 shows a virtual image of the BiPortal
service.
Internet of Bicycles - Tracking and Monitoring Life-cycle Information using GS1
61

Figure 4: BiPortal Installation Example.
BiPortal service is a security service that enables
smartphones to detect a thief and track stolen
bicycles with a beacon installed on a bicycle and a
surveillance device installed on a fixed place such as
a parking facility. Information generated by BiPortal
is saved and shared through Oliot EPCIS.
Figure 5: Elements in BiPortal system.
To build BiPortal, we implement four elements –
BiKeeper (Beacon), BiKeeper (Application),
BiSecure and BiPortal – as described in Figure 5 and
Table 5.
Table 5: Elements and their characteristics.
Elements Characteristics
BiKeeper (Beacon)
- Detect an abnormal behaviour
- Send sensor data
* BiKeeper(B)
BiKeeper
(Application)
- Register BiKeeper (Beacon)
- Set a bicycle’s lock up
- Report a bicycle theft
* BiKeeper(A)
BiSecure
- Collect information from BiKeeper
(Beacon)
- Alert a bicycle theft
- Record ad send video streaming
BiPortal
- Save a video file from BiSecure
- Save information about BiKeeper
(Beacon)
- Manage stolen bicycles’ information
- Save EPCIS events
BiPortal has three layers as shown in Figure 6:
 Service (include BiKeeper(A))
 Server (include BiPortal)
 Hardware (include BiKeeper(B), BiSecure)
The service layer provides a user interface and a
server interface. This layer is implemented as a
mobile application. A mobile application captures
information from sensors and sends information to
the server layer. The server layer, including ONS,
EPCIS and an application server, is for saving and
sharing information.
Figure 6: Service Components.
The hardware layer generates information from
sensors in a beacon and a mobile phone. In this
service, we use a gyroscope sensor in a beacon and a
GPS in a mobile phone. A camera on a RaspberryPi
is used as CCTV for bicycles.
3.1 Oliot IoT Platform
We apply Oliot EPCIS to save and share information
from bicycles. Oliot EPCIS provides web service
interfaces. We install Oliot EPCIS on the
OpenStack-based cloud service to use it.
To provide the BiPortal service, we choose four
bizStep values – private_enrol, robbery_reporting,
riding and parking – from Global Bicycle
Information Architecture.
Table 6: Business step and its description in BiPortal.
bizStep value Description in BiPortal
private_enrol
A beacon is registered to BiPortal
service.
robbery_reporting
A stolen bike is reported to BiPortal
service.
riding An end user rides a bicycle.
parking An end user stops riding.
3.2 BiPortal Service
BiPortal service has two major features, a thief
alarm and stolen bicycle tracking.
To use BiPortal service, a user attaches
BiKeeper(B) to his bicycle. BiKeeper(B) has a
gyroscope sensor to detect abnormal behaviour from
thieves. BiKeeper(B) sends the Received Signal
Strength Indication (RSSI) and gyroscope sensor
IoTBD 2016 - International Conference on Internet of Things and Big Data
62

data to BiKeeper(A) or BiSecure.
After attaching BiKeeper(B), a user who joins
the BiPortal service provides his bicycle and beacon
identification to BiKeeper(A). This information is
used for a user and his bicycle’s specifications.
This application saves login data and a user’s
bicycle information to an application server, and this
information can be saved in EPCIS. The application
obtains event data about riding, parking and robbery
reporting.
BiSecure is installed on a parking facility for
bicycles. When BiSecure obtains information from
nearby bicycles, it will send information to an
application server too. For BiSecure, a video camera
that enables users to watch their bicycles directly is
also installed.
3.2.1 Function 1: Thief Alarm
To detect a theft events, this application has two
features:
 Detect with distance
 Detect with vibration
In a parking state, if an abnormal vibration is
detected, this application notifies the owner with a
display and sound. Or if the bicycle is moved far
from the user in the parking state, this application
also notifies the owner with a display and sound. If a
bicycle is stolen, the owner can record the bicycle as
stolen state.
The average cycling speed is 19.3km/h according
to Road-bike.co.uk (Road-bike.co.uk, n.d.). This
means a thief can move about 320 meters within a
minute on a stolen bicycle. Hence, this application
alerts users at distance of less than 50 meters so that
it will be possible to reach a thief before he or she
escapes.
3.2.2 Distance based on a Beacon
RSSI is the general metric used to define the
distance between a beacon and its observing device
– BiSecure or BiKeeper(A) in this case. It is based
on the TXPower level. Because of its characteristics,
the distance is not accurate when there are any
obstacles nearby.
Figure 7: RSSI Windows.
To increase the accuracy, this application
suggests ‘windows’ logic. The application uses ten
windows to obtain RSSI data and calculate an
arithmetic mean. Figure 7 shows the suggested
windows logic.
There is no abnormal data removal because it is
not possible to calculate the mean if an abnormal
RSSI value is removed from a beacon. Given
TXPower and RSSI values are transformed to
distance based on the following code from Android
Beacon Library (Android Beacon Library, n.d.):
double ratio = rssi*1.0/txPower;
if (ratio < 1.0) {
return Math.pow(ratio,10);
}
else {
double accuracy =
(0.89976)*Math.pow(ratio,7.7095) + 0.111;
return accuracy;
}
3.2.3 Function 2: Stolen Bicycle Tracking
People who use this service are able to track a stolen
bicycle’s location. If a stolen bicycle is observed,
this application gives location information to the
stolen bicycle’s owner.
Because of a privacy issue, this application only
tracks GPS location and a bicycle identifier. The
bicycle identifier and the owner information are
invisible to other users.
3.3 Demonstration
A user who uses BiPortal service can receive an
alarm and a screenshot as in Figure 8. The picture on
the left shows a ‘steal catch’ alarm when abnormal
behaviour is detected from the bicycle. The picture
on the right shows live video from a camera attached
on BiSecure.
Figure 8: BiPortal Application Example.
Internet of Bicycles - Tracking and Monitoring Life-cycle Information using GS1
63

4 CONCLUSIONS
Global Bicycle Information Architecture based on
GS1 provides connections between stakeholders.
These connections become the basis of information
sharing. This idea also enables services that can
protect people’s property and help governments
manage public bicycle systems or appropriately
allocate budget to build bicycle roads and bicycle
parking facilities.
Global Bicycle Information Architecture also
enables industries to find opportunities for providing
various services such as the bicycle portal service
described in this research.
This research suggests a merged identification
system that uses SGTIN and manufacturer’s serial
number together to identify each object and retain
the manufacturer’s original traceability. In addition,
it enables governments to trace recalled bicycles
easily using GTIN.
Based on this research, stakeholders related to
bicycle industries should expand Global Bicycle
Information Architecture and find vocabularies
which are not defined in this research.
REFERENCES
Android Beacon Library, n.d. Available from:
<http://www.github.com/AltBeacon/android-beacon-
library>.
Byun, J., Kim, D., 2015. Oliot EPCIS: New EPC
information service and challenges towards the
Internet of Things, RFID (RFID), 2015 IEEE
International Conference on: pp70-77.
Cespedes, S., Salamanca, J., Sacanamboy, J.C., Rivera, C.,
2014. Poster: Smart networked bicycles with platoon
cooperation, Vehicular Networking Conference
(VNC), 2014 IEEE on: pp.199-200.
Dezani, H., 2015. Urban Road Optimization Based on
Safety and Cyclists' Effort Required by Bike Tracks,
Intelligent Transportation Systems (ITSC), 2015 IEEE
18th International Conference on. IEEE, pp.1111-
1116.
Erlanger, S., 2009. French Ideal of Bicycle-Sharing Meets
Reality, The New York Times. Available from: <
http://www.nytimes.com/2009/10/31/world/europe/31
bikes.html>.
Gartner, 2013. Gartner Says the Internet of Things
Installed Base Will Grow to 26 Billion Units By 2020.
Gartner. Available from: <
http://www.gartner.com/newsroom/id/2636073>.
GS1, 2014. Core Business Vocabulary, GS1, Version 1.1.
Krissoff, Barry, et al., 2004. Traceability in the US food
supply: economic theory and industry studies,
Washington, DC: US Department of Agriculture,
Economic Research Service: pp3-10.
Moss, J., 2014. Data table of thefts per force over the last
5 years, Stolen Bikes UK. Available from: <
https://stolen-bikes.co.uk/statistics/>.
Road-bike.co.uk, n.d.. Average cycling speed for new and
experienced cyclists, Road-bick.co.uk. Available from:
<http://www.road-bike.co.uk/articles/average-
speed.php>.
Russell, M., 2015. The Bike Sharing World - 2014 -Year
End Data, The Bike-sharing Blog. Available from: <
http://bike-sharing.blogspot.kr/2015/01/the-bike-
sharing-world-2014-year-end.html>.
Sundmaeker, H., Guillemin, P., Friess, P., Woelfflé, S.,
2010. Vision and challenges for realising the Internet
of Things, European Commission.
Telegraph Men, 2014. Can this intelligent bike really save
you from crashing?, The Telegraph. Available from: <
http://www.telegraph.co.uk/men/active/recreational-
cycling/11303207/Can-this-intelligent-bike-really-
save-you-from-crashing.html>.
IoTBD 2016 - International Conference on Internet of Things and Big Data
64