Abstract :
[en] In the world of short and medium-range wireless technologies, Bluetooth has recently come to the
forefront of innovation. Within the next five years its market presence, especially in its Low Energy
variation, is expected to nearly double across all market segments. The technology is quickly and
steadily gaining importance for a wide range of applications with a specific focus on Internet of Things
(IoT) devices. The growing availability and variety of such devices constitute an untapped potential
that we plan on exploiting. Our focus in this thesis is to understand Bluetooth’s capabilities and
explore its potential in mobile contexts. One specific field where this technology remains unexplored
is Vehicular Ad Hoc Networks (VANETs). Because of the need to implement and moderate vehicular
communications, the topic of Intelligent Transportation Systems (ITSs) is now trending more than
ever.
In this thesis we propose two ways we can benefit from Bluetooth in a mobile environment. Firstly,
we consider the technology as a communication medium to investigate how di↵erent mobilities a↵ect
the link performance between two devices. To do this, we define a set of communication experiments,
in our case between two vehicles, to analyse how Bluetooth Low Energy (BLE) is a↵ected by varying
speed, distance and traffic conditions. We find that the maximum communication range between two
devices can go beyond 100m and that a robust connection, capable of handling sudden signal losses
or interference, can be achieved up to a distance of 50m. The experiments were conducted using a
proof-of-concept mobile application for o↵-the-shelf smartphones that can be used to transmit data
over multiple hops in various Vehicle-to-Everything (V2X) scenarios.
Secondly, we consider Bluetooth discovery capabilities as an information medium by using a connectionless
approach to analyse di↵erent mobility frameworks. As there is an increasing need for
vehicles and objects to become aware of their context, we implement Bluetooth as a sensing system to
provide contextual information about its surroundings. Our challenge is to find out to what extent we
can exploit the Bluetooth discovery and beaconing scheme for this purpose. We collect and analyse
a dataset of Bluetooth Classic and BLE discoveries and evaluate their respective characteristics and
ability to provide context-aware information from a vehicular perspective. By examining data recorded
about encountered devices, such as quantity, quality of signal and device class information, we infer
distinctive Bluetooth behaviours related to context and application. For this purpose, we propose a
set a features to train a classification model to recognize di↵erent driving environments (i.e. road
classes). Investigating the performance of our classifier, we were able to predict up to three classes
(highway, city, extra-urban) by using only Bluetooth discovery data and no geographical information.
This outcome gives promising results targeted at low energy and privacy-friendly applications and can
open up a wide range of research directions.
In conclusion, in this thesis we present two ways of applying Bluetooth to mobile contexts for
deploying novel human mobility applications.
