IoT based Proximity Marketing
Zanele Nicole Mndebele and Muthu Ramachandran
School of Computing, Creative Technologies and Engineering, Leeds Beckett University, Leeds, U.K.
Keywords: IoT, GCM, Google Cloud Messaging, Wi-Fi, Proximity Marketing, PMaaS.
Abstract: Modern communication is moving toward a digital paradigm influenced by increasing connectivity and the
IoT. Digital communication can be improved by applying proximity rules to improve relevance especially
for marketing messages. The objective of this study was to demonstrate how cloud based proximity
marketing can be implemented as a service on existing wireless connectivity service platforms to deliver
messages that are timely and relevant, using Wi-Fi broadcasts. Information about networking technologies
and proximity determination was used to develop a prototype proximity marketing system to demonstrate
the concepts of Proximity Marketing as a Service that can run on a wireless network. The prototype system
Precinct PMaaS was successfully designed, implemented and tested. When compared to similar Bluetooth
tools the cloud based WiFi driven Precinct PMaaS solution proved to be more efficient and effective,
offering a better value proposition than Bluetooth proximity marketing tools. This study demonstrates how
to achieve proximity communication cost effectively using network service information, demonstrated in a
Wi-Fi only environment. This is ground work on which future projects can apply Big Data analytics to
improve impact of proximity driven digital marketing.
1 INTRODUCTION
Today’s society thrives on mobile connectivity and
the convenience of communication on the move
thanks to the internet of things (IoT). For this reason,
many organizations use digital marketing platforms
to connect to their customers. Digital marketing is
when an organization uses websites, email, texts,
mobile applications (mobile apps) and multimedia to
communicate with customers. Digital marketing has
gained popularity in proportion with increased
market penetration of mobile devices, smart phone
technology and mobile wireless internet services,
supported by ever improving IoT technologies
(Smith, 2011; Hoehle and Venkatesh, 2015)
Digital marketing can be augmented by applying
proximity rules to target the appropriate audience at
the right time and in a specified location. The term
proximity refers to spatial displacement, i.e. how
close one object is to another, or how close an object
is to a certain point or destination (Hightower and
Borriello, 2001). Proximity marketing is a form of
targeted digital marketing that facilitates
communication to a specific audience in a certain
location or within a predetermined range of a
marketing broadcast signal. This type of marketing
is less intrusive than non proximity related
communication, as it confines the communications
to certain time and space to improve message timing
and relevance for a specified target audience
(Kurkovsky and Harihar, 2006).
This study explored the requirements for
enabling IoT based proximity communication, in a
cost effective manner using network information.
The main deliverable was the development of
Precinct PMaaS, a proximity communication system
for broadcast messaging. This study was based in
Leeds city, UK. In an urban setting like Leeds city,
where free Wi-Fi connectivity is offered as a value-
add service in many locations, proximity messaging
would be a viable form of customer engagement.
2 BACKGROUND
Digital marketing is an attractive marketing channel
is because it is relatively less expensive than other
forms of advertising, and can give organizations
unlimited personalised access to their customers,
allowing them to send customized information right
into each customer’s preferred choice of
communication device (Rettie et al., 2005; Shankar
Mndebele, Z. and Ramachandran, M.
IoT based Proximity Marketing .
DOI: 10.5220/0006347903250330
In Proceedings of the 2nd International Conference on Internet of Things, Big Data and Security (IoTBDS 2017), pages 325-330
ISBN: 978-989-758-245-5
Copyright © 2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
325

et al, 2010). However, this type of communication
can become intrusive if it is not used appropriately,
resulting in complaints regarding excessive
marketing communication, lack of relevance and
poor timing of the messages (Smith, 2011;
O’Mahony 2012; Rogers 2015, Information
Commissioner’s Office, 2016)
Proximity related communication is a logical
next step for businesses especially, if they already
have a mobile app or e-commerce presence in the
digital marketing space, as it addresses some of the
issues relating to messages relevance and timing
effectively (Kurkovsky and Harihar, 2006; Krum,
2010).
Three technologies; namely Bluetooth, Wi-Fi
and RFID already have a strong presence in the
digital marketing space. Initial investigations into
proximity marketing indicated that proximity driven
marketing systems have predominantly been
developed with the use of Bluetooth beacons
(Kurkovsky and Harihar, 2004; Krum, 2010). The
burning question was whether it would be possible
to provide simpler solution based on IoT and
focussing on the prevalent Wi-Fi connectivity
around Leeds. Wi-Fi was also appealing because it is
the most prevalent form of wireless connectivity,
provides fast connectivity and coverage over large
areas (Friedman et al, 2013), and is commonly
offered as a value-add internet connectivity service
that people are already accustomed to accessing.
The potential for a solution based on wireless
network information was investigated and as a result
this study proposes the concept of Proximity
Marketing as a Service, PMaaS for short. In PMaaS,
proximity communication is a service that runs on
an existing network and uses information about the
network to deliver proximity driven communication.
A prototype proximity marketing system Precinct
PMaaS was created to demonstrate the concept.
3 PROXIMITY MARKETING AS
A SERVICE (PMAAS)
Precinct PMaaS was envisioned as a system that
would target a particular community of subscribers
in a specific location, and deliver messages to their
mobile devices, via wireless connectivity technology
as long as they remained in the specified location
and only while they were connected to a particular
network segment or specific access point. The
service framework diagram below shows the
principle services that make up PMaaS.
Figure 1: PMaaS Service Framework Diagram.
3.1 System Architecture
As shown in figure 2 below, Precinct PMaaS uses
network information, i.e. the network name, and
GPS parameters derived from the position of the
mobile device to verify if the subscriber should
receive proximity driven communication.
The network name is the primary criteria used to
determine if the subscriber is connected to the
proximity communication network. This is done by
identifying the active network name or service set
identifier, SSID, but could also be done by
determining the access point name or basic service
set identifier, BSSID, (Juniper Networks, 2013).
Precinct PMaaS makes use GPS parameters to
verify the physical location of the subscriber. This is
achieved through GPS, A-GPS (Assisted GPS), Wi-
Fi positioning or cellular network positioning
(Hightower and Borriello, 2001; Zandbergen, 2009;
North, 2011). Nowadays many mobile devices are
equipped location detection functions that are part of
most mobile smart phone operating systems and use
one or more of these technologies to determine
device location.
Wireless broadcast messages are sent by a cloud
based communications platform and a mobile app
installed on the mobile phone listens for messages
applicable to the subscriber, while the subscriber is
in the relevant proximity. See figure 2 below, the
Precinct PMaaS Architecture.
IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security
326

Figure 2: Precinct PMaaS Architecture.
4 DESIGN & IMPLEMENTATION
4.1 Implemented Design
Precinct PMaaS consists of the following
components:
• Data Storage
The database holds two tables; one table stores user
information and communication preferences and the
other table stores a history of messages sent by the
system, as well as the outcome of the message
delivery attempts.
• Mobile Application
An Android mobile application was developed to
cater for interfacing with the subscriber via their
mobile phone. The subscriber uses the app to
register and subscribe to notifications, and the app is
also used to deliver messages to subscribers.
• Messaging Service
A cloud based communications platform, Google
Cloud Messaging (GCM), is used for sending
messages to mobile devices. The Mobile app
connects to GMC via an API (Android Developer,
2016). GCM issues device registration ids, on
registration of the app, which are later used to
identify devices to which message broadcasts should
be delivered. If the application is installed but the
user registration step is not completed, the device
will not receive messages.
• Web Interfaces
Web interfaces facilitate the creation of broadcast
messages and act as a link between Precinct PMaaS
and GCM. Messages are submitted on the web
interface and outcomes of message delivery attempts
are written to a delivery confirmation web interface.
• Location Verification Engine
A Location Verification engine was built into the
mobile application and is used to determine the
subscriber’s location. The engine uses the network
name and the most up-to-date device positioning
information, to determine if the subscriber is
connected to the proximity communication network
and also currently in the correct location to receive
proximity related messages.
4.2 Implementation Process
Precinct PMaaS is subscription driven to avoid being
intrusive. The subscriber must install the Precinct
PMaaS mobile application and register for the
communication categories that are of interest to
them. One registered, the mobile app will only
receive message broadcasts from the subscribed
categories when the subscriber is in the correct
location and connected to the correct network.
IoT based Proximity Marketing
327

Figure 3: System Components.
5 SYSTEM TESTING AND
EVALUATION
Precinct PMaaS was tested and also compared to
similar Bluetooth solutions. All tests were conducted
in the Student Hub at Leeds Beckett University
Headingley Campus. Three mobile phone devices
were used for testing; a Huawei Y360-U31, an
Alcatel One Touch PIXI 3, and a Sony Xperia J.
Findings are discussed in the section 6.
5.1 Unit Testing
Precinct PMaaS was developed as a modular system.
All components and functionality were tested
iteratively after each module was developed and
added to the core. Then the final system was tested
end to end and evaluated by end users.
5.2 Location Verification Testing
The mobile app was configured to show an alert that
would confirm if the mobile phone was connected to
the proximity network and was in the right location,
as determined through GPS coordinates. All three
mobile phones were used in testing the location
verification engine. The Alcatel mobile phone was
configured to produce negative results by altering
values for expected GPS coordinates and expected
network name alternately. When tests were run,
broadcast messages were successfully delivered to
the Huawei and Sony mobile devices. Alerts on the
Alcatel indicated that it was not in the correct
proximity; proving that the location verification
engine worked.
5.3 Message Delivery Efficiency Tests
on Bluetooth Software
Three free downloadable Bluetooth proximity
marketing software tools, AreaBluetooth, FreeBlue,
and Blue Magnet, were installed and tested to see
how long it took to deliver messages. The aim was
to explore these solutions and use the test results as a
baseline to assess performance of Precinct PMaaS.
Message delivery was measured on a stopwatch
for Bluetooth tool tests as well as for Precinct
PMaaS tests in order to maintain testing consistency.
Ten sets of test were run for each Bluetooth device
over a course of 3 days, at a distance under 10
meters from the Bluetooth dongle.
5.4 Message Delivery Efficiency Tests
on Precinct PMaaS
For Precinct PMaaS, three sets of tests were run at
IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security
328

each of four distances from the access point, for 3
consecutive days. The tests were conducted at
distances of 0 to 1 meters, 6.5 meters, 10 meters and
16.5 meters away from the access point. The Alcatel
device was selected as the primary test device from
which network signal strength was checked and
recorded just before each test was run. All three
mobile phones were included in the testing in order
verify results through simultaneous message
delivery.
5.5 End User Evaluation
The Precinct PMaaS system was demonstrated to a
broad audience. The system was also evaluated by
users who were satisfied that the basic principles of
proximity marketing were addressed.
6 FINDINGS AND DATA
ANALYSIS
6.1 Findings on System Performance
Results and findings from Precinct PMaaS and
Bluetooth message delivery tests were as follows:
• On average messages were delivered faster over
Precinct PMaaS than over Bluetooth solutions
in equidistant tests and they arrived on all three
devices simultaneously, see table 2.
• Bluetooth solutions needed to detect the device
first before transmission; device detection took
on average 10 seconds. The detection times
were excluded from the test results.
• Bluetooth solutions had a shortcoming in that
they indicated that they could only connect to a
maximum number of seven devices at a time.
• Overall Precinct PMaaS offers better distance
coverage, performance and capacity than the
freeware Bluetooth solutions that were tested.
Table 1: Average Precinct PMaaS Message Delivery
Times (in seconds) by Distance.
Distance
Day 1
Averages
Day 2
Averages
Day 3
Averages Average
0 to 1
Meter 1.29 0.87 1.15 1.10
6.25
Meters 1.18 1.07 1.25 1.17
10
Meters 1.45 1.04 1.20 1.23
16.25
Meters 1.98 1.23 1.15 1.45
Overall Average : 1.24
Table 2: Comparison of Bluetooth and Precinct PMaaS
text message delivery times.
Comparison of Bluetooth and Wi-Fi Text Message
delivery times: max 10 meters from signal origin
(10 tests per Bluetooth system)
System Type
Average time
in seconds
Blue Magnet Bluetooth 5.40
AreaBluetooth Bluetooth 3.00
FreeBlue Bluetooth 9.79
Precinct PMaaS Wi-Fi 1.23
6.2 Statistical Analysis
Basic statistical analysis on data from the Precinct
PMaaS messages delivery tests produced the
following findings:
• Fastest message delivery time recorded was 0.8
seconds, and the slowest was 3.0 seconds
• Lowest signal strength recorded was -79dBm
and the highest was -53dBm.
• The most frequent signal strength was -66dBm,
and the most frequently recorded message
delivery time was 1.13 seconds.
• Based on this data set, a message delivery time
of 1.24 seconds is a realistic expectation for an
average signal strength of -63dBm.
• The standard deviation figures were relatively
low; which implies the data is reliable
Figure 4: Basic Statistical Measures.
7 CONCLUSIONS
Precinct PMaaS was developed based on literature
about the IoT in four areas; wireless technology,
location detection, proximity marketing and message
broadcasting. This study found that cloud based
Proximity marketing running as a service driven by
network information is not only feasible, but offers
better value than Bluetooth in terms of efficiency
and cost effectiveness, as it can be used to
implement IoT based proximity solutions that run on
existing infrastructure. IoT based Proximity
IoT based Proximity Marketing
329

communication requires the following key
information to work:
• The ability to identify and retrieve network
service information.
• The ability to uniquely identify a device or
subscriber that should receive the
communications.
• The ability to determine if the device or
subscriber is indeed connected the network
associated with the proximity communications.
When configuring proximity rules, combining the
SSID and GPS is recommended for a network with
multiple access points at one site or a network with
multiple sites and multiple access points. For a
network with one site and one access point, a rule
that checks the SSID or the BSSID would suffice.
7.1 Future Work
The nature of proximity marketing opens up many
possibilities in terms of understanding how to track
consumer habits and the impact of digital marketing
messages. Information derived from digital
footprints of mobile connectivity proximity driven
by IoT can potentially provide insights to inform
future mobile communications development. Future
research should explore how Big Data can provide
insights relating to consumer reactions to mobile
proximity driven communication, and possibly
contribute to future development of predictive and
proactive proximity communication (Want and
Schilit, 2001; Lee and Gnawali, 2015).
ACKNOWLEDGEMENTS
I would like to acknowledge Dr Mario Marino, from
Leeds Beckett University, for his contributions,
guidance and insightful feedback in the production
of the dissertation on which this study is based.
REFERENCES
Android Developer (2016) Cloud Messaging. Android
Developer [Online] Available from:
<https://developers.google.com/> [Accessed 15 July
2016].
Friedman, R. Kogan, A. Krivolapov, Y. (2013) On Power
and Throughput Tradeoffs of Wi-Fi and Bluetooth in
Smart phones. IEEE Transactions on Mobile
Computing. 12 (7), pp. 1363- 1376.
Hightower, J. and Borriello, G. (2001) Location Systems
for Ubiquitous Computing. Computer, 34(8) August,
pp. 57-66.
Hoehle, H. and Venkatesh, V. (2015) Mobile Application
Usability; Conceptualization and Instrument
Development. MIS Quarterly, 39(2), pp. 435-472.
Information Commissioner’s Office (2016) Guide to
Privacy and Electronic Communications Regulations
[Online]. Available from Available from:
<http://www. ico.org.uk/> [Accessed 28 May 2016].
Juniper Networks (2013) Understanding the Network
Terms SSID, BSSID and ESSID. Juniper Networks
[Online]. Available from: <https://www.juniper.net>
[Accessed 06 June 2016].
Krum, C. (2010) Mobile Marketing: Finding Your
Customers No Matter Where They Are. Pearson
Education, USA.
Kurkovsky, S. and Harihar, K. (2006) Using Ubiquitous
Computing in Interactive Mobile Marketing. Personal
and Ubiquitous Computing, 10 (4) May, pp 227–240.
Lee, H.J. and Gnawali, O. (2015) Predictive Data Delivery
to Mobile Users Through Mobility Learning in
Wireless Sensor Networks. IEEE Transactions on
Vehicular Technology, 64 (12), pp 5831 - 5849.
North, M.M. (2011) An Empirical Analysis Of Mobile
Smart Device Accuracy and Efficiency in GPS-
enabled Field Data Collection. Issues in Information
Systems, 13 (1), pp 318 - 327.
O’Mahony, J. (2012) Spam text pair face £250000 fine.
Telegraph UK [Online], 1 October. Available from: <
http://www.telegraph.co.uk> [Accessed 16 May
2016].
Rettie R., Grandcolas, U. and Deakins, B. (2005) Text
message advertising: Response Rates and Branding
Effects. Journal of Targeting, Measurement and
Analysis for Marketing, 13(4), pp. 304-312.
Rogers, K. (2015) LinkedIn Notorious for Sending Too
Many Emails Cuts Back. New York Times [Online], 29
July. Available from :< http://www.nytimes.com/>
[Accessed 05 May 2016].
Shankar, V. Venkatesh, A. Hofacker, C. and Naik, P.
(2010) Mobile Marketing in the Retailing
Environment: Current Insights and Future Research
Avenues. Journal of Interactive Marketing, 24, pp.
111–120.
Smith, K. T. (2011) Digital Marketing Strategies that
Millennials Find Appealing, Motivating, or Just
Annoying. Journal of Strategic Marketing, 19 (6)
October, pp. 489-499.
Want, R. and Schilit, B. (2001) Expanding the Horizons of
Location-Aware Computing. Computer, pp. 31-34.
Zandbergen, P. (2009) Accuracy of iPhone Locations: A
Comparison of Assisted GPS, Wi-Fi and Cellular
Positioning. Transactions in GIS, 2009, 13, pp 5 - 26.
IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security
330