Journal of Global Positioning Systems (2004)
Vol. 3, No. 1-2: 200-207
Secure Tracking using Trusted GNSS Receivers and Galileo
Authentication Services
Oscar Pozzobon1, Chris Wullems1, Kurt Kubik2
1 Qascom, Via O.Marinali 87, 36061 Bassano del Grappa (VI), Italy
e-mail: [e-mail], [e-mail]; Tel: + 39 0424-525-473; Fax: [phone]
2 University of Queensland, Brisbane QLD, Australia
e-mail: [e-mail]; Tel: [phone]; Fax: [phone]
Received: 08 December 2004 / Accepted: 03 February 2005
Abstract. This paper describes a secure framework for
tracking applications that use the Galileo signal
authentication services. First a number of limitations that
affect the trust of critical tracking applications, even in
presence of authenticated GNSS signals, are detailed.
Requirements for secure tracking are then introduced;
detailing how the integrity characteristics of the Galileo
authentication could enhance the security of active
tracking applications. This paper concludes with a
discussion of our existing tracking technology using a
Siemens TC45 GSM/GPRS module and future
development utilizing our previously proposed trusted
GNSS receiver.
Key words: Galileo, GPS, security, authentication,
privacy, tracking
1 Introduction
In recent years there has been an increasing presence of
real time tracking applications in the GNSS (Global
Navigation Satellite Systems) market. It is possibly one
of the fastest growing areas in GNSS, however, fast
growth brings with it significant drawbacks such as the
lack of standardization and security support for
applications that require security for revenue protection
or even safety critical services.
The emergence of low cost GSM GPRS (General Packet
Radio Service) providing wireless communications
throughout Europe has fuelled rapid growth in the
development of embedded, application specific platforms,
which provides low-cost telematic solutions that are easy
to integrate. This rapid growth has happened at the
expense of a suitable security framework for development
of safety-critical or financially-critical applications. To
date there are few telematic solutions on the market that
even offer security services. Most rely on the flawed
encryption and key establishment protocols of GSM. One
of the most significant issues with GSM encryption is the
lack of support for cryptographic integrity protection.
The increase of tracking applications has resulted in the
emergence of a number of new applications including
GPS based road toll payments, GPS based insurance
policy, location-based access control, finance and
tracking for security. A number of these applications will
be introduced in the following sections.
1.1 GNSS based road toll collection
Future road toll collection systems are planned to be
based on GNSS technology in order to reduce
infrastructure costs and to achieve region based tax
collection facilitating regional tolling variations such as
pollution-tax for highly polluted areas. As taxes are
calculated based on location data from a GNSS receiver,
it is imperative that location data is trusted. An additional
concern is management of privacy in these technologies.
This issue is to a large extent sill unresolved, as location
data obtained by operators of these schemes are hardly
controlled by the user.
In mid-2003, the EC (European Commission) adopted a
proposal (COM(2003) 488) to align the national systems
of road tolling for heavy-goods vehicles in Europe. This
directive does not impose a particular technical solution,
but another EC communication (COM(2003) 123)
proposes the use of GNSS positioning and GSM/GPRS
mobile communications for new electronic toll systems
from 2008 onward (for all vehicle types). In addition, it
proposes that migration from legacy microwave systems

Pozzobon et al.: Secure Tracking using Galileo 201
to GNSS / mobile communications should be complete
by 2012. Authentication, availability and integrity of the
location data will be the main technical problem for
revenue protection of toll operators. From the user
perspective, privacy continues to be the biggest concern,
which will demand innovative solutions for the
management of location data privacy.
1.2 GNSS based insurance policy
GNSS tracking systems can be applied to insurance in
obtaining time usage and location travelled by insured
vehicles. This will permit insurance companies to create
pay per usage policies. For some users this would result
in cost reduction, as the policy cost can be estimated by
vehicle usage as total kilometres, risks of travelled
regions, average speed, time usage and time based risk.
The use of this type of information in insurance policies
continues to be an active area of research. A number of
experiments have been conducted by Norwich Union and
IBM in this area to test the user responses and system
reliability.
1.3 GNSS based aircraft tracking
The Future Air Navigation System (FANS) is an example
of aircraft tracking using a system facilitating a free
flight. In this technology, FANS is responsible for
communications, navigation and surveillance. The Flight
Management Computer (FMC) uses GPS, inertial
measurements, air data, and other navigation radios if
available to facilitate surveillance by a traffic
management (ATM) center, such that aircraft can be
tracked at all times. The security of this system is
important where there is an absence of radar systems to
verify the reported location of the aircraft. Security of the
data transmitted to the ATM is particularly critical in this
situation, as it can assist in preventing intentional location
spoofing.
1.4 GNSS based access control and auditing
Location can be used for the enhancement of access
control systems and auditing. There are many
applications where location context information can
supplement existing security. Such applications would
utilize location from GNSS systems in providing
location-based audit trails, or access control policy where
resources are granted or denied based on the location of a
user. Security and trust of location is particularly
important in this type of application, as insecure location
acquisition would not only result in serious security
breaches, but a false sense of security.
2 Limitations of existing technologies
There a number of security issues with tracking systems
which utilize existing GNSS and telematic devices. These
issues limit the potential for development of critical
applications. The most significant limitations are:
1. Lack of signal authentication: There is no civil
method to authenticate the GPS signal;
2. Lack of framework or standardized methodology
to verify the integrity of a device and assess its
security;
3. Lack of standardized telematic protocols that
provide communications security; and
4. Significant privacy issues, such that it is difficult
to obtain the location and preserve privacy at the
same time.
The following subsections discuss these limitations in
terms of signal authentication, device integrity and
telematic security.
2.1 Signal authentication
There have been a number of recent efforts to quantify
the extent of vulnerabilities and limitations the GPS civil
signal imposes on civil applications in the presence of
malicious attacks. Perhaps the most prominent
vulnerability analysis was the report on the vulnerabilities
of GPS in transportation, performed by the Volpe center
for the US Department of Transportation (Volpe 2001).
As the GPS civil signal is not authenticated, it is possible
to simulate it. In recent years simulators have become
readily available, such that a GPS simulator can be hired
relatively cheaply and can be connected to the antenna of
a GPS receiver in a vehicular tracking module for
example. Because of the possibility of signal simulation,
the current generation of GPS tracking modules poses a
potential security risk for use in safety or financially
critical applications, such as the tracking of hazardous
materials.
The U.S. Senate has recently approved a measure
providing funds for the Transportation Security
Administration (TSA) to develop measures for tracking
trucks carrying hazardous materials (HAZMAT). As
stated by (Gibbons 2004) More than 800,000 shipments
of hazardous materials take place in the United States
every day, including flammable fuel products, potentially
explosive fertilizers, and volatile chemicals. GPS signal
authentication is necessary for secure tracking in order to
prevent a hijacker from simply spoofing the reported
location.
The risk of signal simulation attacks is significantly
reduced where the cost of hiring a signal simulator

202 Journal of Global Positioning Systems
outweighs the potential cost savings in defeating a
tracking system. An example of this is performing a
simulation attack in order to avoid the payment of road
tolls or to cheat an insurance company by falsely
reporting the number of kilometers traveled.
In addition to GPS signal spoofing, there is the potential
for spoofing of augmentation data. The most widely used
augmentation systems are satellite based augmentation
systems including the European Geostationary
Navigation Overlay Service (EGNOS) and the American
Wide Area Augmentation System (WAAS). Both these
augmentation systems do not provide authentication or
cryptographic integrity protection. Spoofing of the
correction data provided by these systems can introduce
small but significant errors, which may be problematic
where a few meters of error is critical.
2.2 Device integrity
A typical tracking device is composed of a GPS module
and a communications module such as a GPRS modem or
radio modem. Some tracking devices additionally contain
a microprocessor with software to process the data from
the GPS module. There are a number of limitations of
current GPS modules:
• There is no cryptographic authentication or
integrity protection of NMEA position, time or
velocity data sourced from the GPS module; and
• The NMEA location data from the GPS module
can be trivially simulated.
In addition to the limitations of GPS modules, the lack of
authentication and end-to-end communications security
between the tracking device and the telematic server can
exacerbate the possible attacks that can be performed,
such as masquerading as the tracking device to spoof the
location.
2.3 Privacy
The growth of GNSS tracking applications has also
created significant privacy concerns. The first techniques
for managing privacy where proposed by (Spreitzer and
Theimer 1993). These techniques were based on a
location broker residing in the middleware layer. In
recent times, considerable research has been conducted in
the specification of protocols and policy representations
in the context of a cellular location. For tracking
applications, privacy is an issue for both single purpose
tracking devices and multipurpose ones. For single
purpose tracking, a user may be concerned with the usage
of the location data by the destination application. Where
a tracking device is used in a multi-application context,
such as a device supporting both toll-payment and
insurance tracking applications, the user must be able to
configure privacy policies such that only the absolute
minimum required location data for a given application is
provided.
The user in effect should be able to adjust the accuracy of
the location observations depending on the intended use
and identity of the recipient, and the compliance with
user’s privacy policy. The most prominent effort to
create location privacy standards outside the cellular
domain is seen in the Geographic Location/Privacy
(Geopriv1) working group of the Internet Engineering
Task Force (IETF). This group has developed a number
of draft standards for representation of privacy policy
data and protocols for management of location privacy.
3 Secure tracking using Galileo
There are a number of requirements for security and
privacy in GNSS tracking applications. The following
subsections discuss these requirements and possible
implementations in terms of satellite navigation system
security, tracking device security, communications
security to the telematic server, and privacy.
3.1 Signal authentication
Signal authentication of satellite navigation systems is
required in order to ensure that the source of the satellite
signaling is not from a simulator, but is genuine. There
are a number of existing and proposed signal
authentication methods (Hein and al 2002), (Scott 2003)
which are summarized below:
• Signal Authentication through Authentication
Navigation Messages (ANM): The ANMs would
include a digital signature authenticating the
other navigation messages that contain data
including ephemeris and almanac data. Using
the digital signature, the certified receiver is able
to authenticate the source of messages and verify
their integrity. These authentication messages
are created on the ground and transmitted to the
satellites for broadcast. This method has a
security limitation, in that the messages can be
acquired by a certified receiver and modulated
over a simulated signal in order to spoof the
Galileo signal. This would require functionality
that is not commonly found in commercial
signal simulators, and would require the
operation to be performed within a very small
time window. Documents from the Galileo
design consolidation indicates that the Galileo
1 Refer to http://ecotroph.net/geopriv

Pozzobon et al.: Secure Tracking using Galileo 203
Open Service may support this type of signal
authentication (Galilei 2003).
• Signal Authentication through Spread Spectrum
Security Codes (SSSC) (Scott 2003): SSSCs are
synchronous cipher streams seeded by an unsent
digital signature from an Authentication
Navigation Message, interleaved with normal
spreading sequences. The receiver stores A/D
samples and once the digital signature is
received, it is able to generate the security
spreading code reference signal and correlate it
with the stored samples. If the SSSC is detected
at the correct power level, the signal is
authenticated. This technique has the innovative
advantage that permits authentication in a signal
open to the public without the difficulties of key
distribution; however it has the limitation that
the spoofing detection is proportional to the
antenna gain and that the authentication
verification is not immediate. A more secure
type of authentication based on SSSCs is also
proposed by (Scott 2003), utilizing a Civil
Antispoof Security Module (CASM) with a
preloaded Red Key and the authentication
navigation message for seeding of the cipher
stream generator. This type of signal
authentication does not have the drawbacks of
the public SSSC version.
• Signal Authentication through Spreading Code
Encryption (SCE): Spreading code encryption is
one of the oldest signal authentication
techniques, currently used by the GPS P(Y)
code, an exclusively military service, and is
projected to provide authentication of the
Galileo CS and PRS signals. As the spreading
code is secret, without knowledge of the
spreading code, signal access is denied. For this
reason, the spoofer cannot simulate the signal,
and hence authentication of the signal is
achieved when the user possesses the correct
spreading code. In GPS’ P(Y) code, the P code
is publicly known, and the secret spreading code
is obtained using P code with a Red Key, or a
Black Key and the Selective Availability Anti-
spoofing Module (SAASM) (Callaghan and
Fruehauf 2003). The Black Key is the Red Key
encrypted with the public key of a given
SAASM. , allowing the Red Key to be decrypted
inside the tamper-resistant SAASM which
contains its private key. The Black Keying
infrastructure allows for electronic key
distribution and does not compromise the
classified Red Key.
Civil signal authentication is a challenge for next
generation satellite systems. As detailed above, there are
a range of different strength security solutions and
proposals. The suitability of a particular signal
authentication mechanism is dependant on the cost to
defeat the mechanism; balanced against the possible gain
should the mechanism be successfully defeated.
The Galileo signals and corresponding authentication
schemes to date have not been decided. Based on
generally available information, it is evident that Galileo
will provide a number of different services, the following
of which are projected to provide signal authentication:
(Hein and al 2002)
• Open Service (OS): Based on available literature
(Galilei 2003), encryption may be provided to
the open service on the E5b-I data channel. This
service will provide authentication through
Authentication Navigation Messages (ANM);
• Safety of Life Service (SOL): This service will
provide satellite and signal integrity messages
and Authentication Navigation Messages
(ANM);
• Commercial Service (CS): This service will
provide access restriction and authentication
through spreading code and data encryption
(SCE); and
• Public Regulated Service (PRS): This service
will provide access restriction and authentication
through spreading code and data encryption
(SCE).
Of particular interest to consumer applications such as
insurance tracking and toll collection is the signal
authentication provided on the Open Service. Fig. 1
illustrates a candidate navigation message authentication
scheme as detailed in the Galilei Project Galileo Design
Consolidation (Galilei 2003).
The authenticated navigation messages would be created
by the Ground Control Centre and up-linked to the
satellites. In theory a public key certificate certified by
the Galileo certification authority would be included in
the navigation messages, and could be verified by the
Galileo certification authority certificate stored on a
certified receiver. Once the public key is verified, the
receiver would be able to verify the signature included in
the navigation messages, and hence authenticate the
source of the navigation messages, and implicitly the
integrity of the messages.
For applications with greater security requirements, the
CS and PRS signals will provide signal authentication
through encrypted ranging codes. While there is no
literature on the key distribution schemes, it can be
assumed that that the implementation would be similar to
the declassified Black Key distribution framework used

204 Journal of Global Positioning Systems
with the P(Y) code of GPS. The key distribution and key
storage problem in this scenario are similar to the
Selective Availability Anti Spoofing Module (SAASM)
(Callaghan and Fruehauf 2003), used in military
applications.
Fig. 1 Navigation authentication message
3.2 Receiver security for tracking applications
We have previously proposed the design for a trusted
civil receiver in (Pozzobon et al. 2004). In this
architecture a tamper resistant receiver uses public key
cryptography to assure the chain of trust to the
application, and provide authentication and cryptographic
integrity of the data to the application. The tamper
resistant receiver acquires and authenticates the signal,
calculates the location and creates data authentication
messages containing a digital signature of the location
data, signal state and tamper-resistance state using the
receiver private key. The data is sent to a telematic server
via a wireless communications service such as GPRS.
The data is send using an extension to the National
Marine Electronic Association (NMEA) protocol 0183
for GPS data navigation we have proposed (Pozzobon et
al. 2004), which provides authenticated position, time and
velocity information as well as signal state and tamper-
resistance state of the receiver.
As integrity verification is a computation embedded in
the device, the computation must be trusted. The device
must implement all the necessary integrity verification
algorithms and transmit to the telematic server any
information regarding status, availability, integrity of the
signal, and the results from the verification operations.
Technology such as the Trusted Computing Platform
(TCP)2 can be used to build trusted systems, that is,
systems where the application running can be trusted,
2 Refer to https://www.trustedcomputinggroup.org/
with the assurance that it has not been compromised, or
modified by an attacker.
Fig. 2 Tracking with a certified receiver
The concept behind this technology is to build a computer
with trusted building blocks (TBB). In TBB, the core root
of trust for measurement (CRTM) is the Basic
Input/Output system (BIOS) of the computer. The CRTM
uses the trusted platform module (TPM) for cryptography
operations (storage of keys, encryption) in order to trust
the system boot3 and verify the integrity of the
subsequently executed applications. The whole “chain of
trust” is based on public key infrastructure (PKI), RSA
and 3DES algorithms.
3.3 Location privacy
As detailed in Section 2.3, there are a set of emerging
standards for location privacy developed by the Geopriv
working group of the IETF. A high level diagram of the
interactions between architecture components in using
Geopriv protocols for management of privacy between a
GNSS tracking device and telematic server is illustrated
in Fig. 3. The components of the architecture include:
• GNSS tracking device: The proposed tamper
resistant GNSS device;
• The Telematic Server: The server that manages
location and communications;
3 The procedure that starts all the necessary process of an
operating system

Pozzobon et al.: Secure Tracking using Galileo 205
• Rule Holder: The entity that provides the rules
associated with a particular target for the
distribution of location information; and
• Rule Maker: The authority that creates rules
governing access to location information for a
target.
Fig. 3 Location privacy using Geopriv protocols
Fig. 3 additionally illustrated the process of privacy rule
upload on the telematic server and how privacy is
protected in the location acquisition process. This
process, which complies with the RFC3693 requirements,
consists of the following steps:
• Rule Transfer: The Rule Maker sends a Rule to
the Telematic Server containing the privacy
information;
• (1a) Signed Rule: the Rule Maker may write a
Rule and place it in a Public Rule Holder as an
alternative. The Telematic Server can access the
Public Rule Holder to read the signed Rules;
• Location Information Request: The Location
Recipient Application requests location
information to the Telematic Server.
• Locate: The Telematic Server is either
continuously receiving the location data from
the GNSS tracking device or can request updates
on the location. The communication is
encrypted; and
• Filtered Location Information: The Telematic
Server sends the location information to the
Location Recipient Application. The
information may be filtered in order to comply
with the privacy policy and rules defined by the
rule maker.
4 Secure tracking using current and emerging
technology
The following subsections detail current and emerging
technologies that have been developed or are currently
under development.
4.1 Current tracking devices developed by Qascom
A number of tracking solutions have been developed
using commercially available hardware such as the
Siemens TC454 GSM module and SiRF5 GPS Receiver.
Fig. 4 illustrates the components in the Qascom tracking
blackbox. The location information is processed by a Java
applet loaded into the flash memory of the TC45 GSM
module. The java applet is responsible for processing
GPS location, velocity and time data. The information is
then processed according to the requirements of the
application, such as insurance.
The applet signs the resulting information destined for the
telematic server by invoking a sign function using the
SIM Toolkit interface6 of the SIM card. This functionality
requires a SIM card with support for public key
operations accessible through the SIM toolkit interface.
The signature of the data provides cryptographic integrity
protection of the data as well as authenticating the source
of the data. The resulting information and signature are
sent to the telematic server using one of two supported
communication modes: SMS (Short Messaging Service)
messaging or GPRS (General Packet Radio Service).
To ensure the privacy of the data communicated to the
telematic server, appropriate security must be used.
Where SMS is chosen as the mode of communication, no
session cryptography is used. This is for three reasons:
• SMS is a point-to-point delivery service where
the end point is another device on the GSM
network. It is assumed that a destination on the
GSM network cannot easily be spoofed. The
tracking device can be authenticated using the
source MSISDN (Mobile Station International
ISDN Number);
• GSM provides data encipherment using the A5
algorithm from the mobile station to the base
stations. The core network is assumed to be
protected from intruders. Flaws in the encryption
and key establishment protocols of GSM are
overcome by the use of new protocols in 3G
deployments; and
• A significant number of messages would be
required for authentication and establishment of
keys, resulting in a communications protocol
that is too costly.
In this mode, the telematic server must have a mobile
station present on the GSM network.
4 Refer to http://www.siemens-mobile.com
5 Refer to http://www.sirf.com/
6 SIM Application Toolkit (SAT) is defined in GSM
11.14 standard for 2G networks, and 3GPP 31.111 for 3G
networks.

206 Journal of Global Positioning Systems
Fig. 4 Qascom GPS/GSM tracking box
Where GPRS is chosen as the mode of communications,
the processed data and signature must pass through the
GPRS Gateway (GGSN) and over the Internet to the
telematic server as illustrated in Fig. 5. For this reason,
the tracking device must be able to authenticate the
destination telematic server, and establish session keys
with this entity in order to transfer the processed data.
This process is facilitated through a small implementation
of SSL (Secure Sockets Layer) in the Java applet.
This implementation of SSL uses the SIM card to verify
the public key certificate of the telematic server and
perform public key operations for the exchange of keys.
Computation of the pre-master-secret is performed in the
Java applet, using the real-time clock in the GSM module
for generation of random numbers. The pre-master-secret
generated by the applet is then encrypted in the SIM card
using the telematic server’s public key. The master secret
is generated from the pre-master secret as specified in
(Dierks and Allen 1999). The master-secret is the key
used for encrypting the session between the device and
the telematic server. The processed data and signature can
then be transmitted over the established security context.
This mode of communications is significantly more
secure than the SMS mode.
Fig. 5 GNSS tracking using GPRS
While the current solution provides protection from
malicious attacks on the communications between the
tracking device and the telematic server, the solution does
not prevent an attacker from loading malicious firmware
(Java applet) onto the tracking device. As the device has
no mechanism to authenticate the firmware, the malicious
software could generate data that minimizes the amount
payable to an insurance company, for example. The
software would still be able to use the cryptographic
functionality of the SIM card, such that the telematic
server would be unaware of such malicious activity.
In addition, the current GPS implementation provides no
signal authentication or protection from spoofing. Thus, it
is possible to attach a GPS signal simulator to the antenna
and spoof the location.
4.2 Next generation devices being developed by
Qascom
We are currently involved in development of a next
generation tracking device that will fulfill the security
requirements of applications such as insurance tracking.
The proposed tracking device will contain a trusted
GNSS receiver, which will be developed by a consortium
of research institutions and companies including Qascom
in Europe. The trusted GNSS receiver is described by
Pozzobon, Wullems and Kubik in (Pozzobon et al. 2004).
The proposed tracking device will contain a general
purpose processor with tamper-resistant key storage and a
cryptographic coprocessor (trusted platform module). In
addition to supporting secure communications over
GPRS, the tracking device will also provide the facility
for authentication of firmware as well as support for
trusted GNSS positioning. Fig. 6 illustrates the secure
tracking box.
The verification of the data provided by the trusted GNSS
receiver is performed by the authenticated software. This
verification initially involves verification of the public
key of the trusted GNSS receiver, before being able to
verify the digital signatures contained within the
authentication data.
This software not only verifies the GNSS position, time
and velocity data, it additionally processes the data as
required by the application. The resulting data to be sent
to a telematic server contains the processed GNSS data,
signal state and device compliance reports obtained from
the trusted GNSS receiver, the public key certificate of
the tracking device, and a digital signature of this data.
The digital signature is calculated by the cryptographic
co-processor using the private key stored in the tamper-
resistant, secure key storage. The telematic server first
must verify the public key certificate of the tracking
device, after which it is able to verify the data received.

Pozzobon et al.: Secure Tracking using Galileo 207
Fig. 6 Qascom secure tracking box
5 Conclusions
This paper has described a secure framework for tracking
applications that use the Galileo Authentication Services.
Requirements for secure tracking in both consumer and
critical applications were introduced, detailing how the
new signal characteristics of Galileo can be used to
enhance the security of tracking applications.
Requirements in terms of authenticated signaling, device
security and location privacy were introduced. This paper
concluded with a discussion of both existing and future
tracking device developments and detailed the strategies
used to mitigate the security issues in the presence of the
Galileo security differentiators.
References
Callaghan, S., and Fruehauf, H. (2003). SAASM and Direct
P(Y) Signal Acquisition, The Journal of Defense
Software Engineering, 16(6), 12-16.
Dierks, T., and Allen, C. (1999). The TLS Protocol Version
1.0, Network Working Group, Internet Engineering
Task Force, Request for Comments 2246.
Galilei. (2003). The Galilei Project: GALILEO Design
Consolidation, European Commission.
Gibbons, G. (2004). HazMat Keeps on Truckin', GPSWorld
(October), 6.
Hein, G. W. et al. (2002). Status of Galileo Frequency and
Signal Design. Brussels.
Pozzobon, O., Wullems, C., and Kubik, K. Requirements for
Enhancing Trust, Security and Integrity of GNSS
Location Services, Institute of Navigation (ION ), 60th
annual meeting, Dayton, OH, USA.
Scott, L. Anti-Spoofing & Authenticated Signal Architectures
for Civil Navigation Systems, ION GPS, GNSS 2003,
Portland, OR.
Spreitzer, M., and Theimer, M. Providing Location
Information in a Ubiquitous Computing Environment.
Fourteenth ACM Symposium on Operating System
Principles, 270–283.
Volpe, J. A. (2001). Vulnerability Assessment of the
Transportation Infrastructure Relying on the Global
Positioning System.