Designing Wireless Automotive Keys with Rights Sharing Capabilities
on the MSP430 Microcontroller
Bogdan Groza, Tudor Andreica and Pal-Stefan Murvay
Faculty of Automatics and Computers, Politehnica University of Timisoara, Romania
Keywords:
Automotives, RF Keys, MSP430.
Abstract:
We explore the ultra-low-power microcontroller MSP430 from Texas Instruments as potential platform for
developing vehicle keys. Radio frequency (RF) keys are still a relevant research subject as they are a common
target for adversaries while automotive manufacturers show an increased interest in adding new functionalities
to traditional keys while keeping them inexpensive. MSP430 is a low-cost, ultra-low-power, 16-bit capable
microcontroller which can handle some cryptographic primitives that can be further used for designing secure
authentication protocols. In this work we do explore the design and implementation options for a protocol
that can be deployed in a car-sharing scenario where multiple users can share or gain access rights to the same
vehicle. Due to inherent constraints of our platform, we keep the protocol simple and rely on inexpensive
symmetric key primitives while still providing advanced options, e.g., rights sharing capabilities.
1 INTRODUCTION AND
RELATED WORK
Despite their apparent simplicity and a rather long de-
velopment history, car keys are still an interesting re-
search subject. This happens because of several re-
asons. First, breaking car keys remains an attractive
subject for hackers due to the inherent value of cars
and nonetheless of the personal belongings inside the
car. As cars are frequently left unattended in remote
locations they are easily accessible to adversaries. Se-
cond, while increasing security is always possible by
using more expensive technologies (e.g., SoA cryp-
tographic designs, etc.) car manufacturers are fre-
quently trying to cut on production costs. Thus one
needs to achieve security at the lowest possible price.
Third, in the recent years there has been a spectacular
growth in functionalities that are present inside cars,
due to the massive growth of technologies that are
helpful to the driver and also entertaining for passen-
gers. Remote controlling such functionalities beco-
mes an immediate necessity and the vehicle key can
be the link to these functionalities.
In this work we try to answer these necessities by
exploring the MSP430 platform from Texas Instru-
ments as potential platform for the development of
RF vehicle keys. The main motivation comes from
the fact that MSP430 is a low-cost, ultra-low power
device that is widely available on the market. Nonet-
heless, it is a capable 16-bit controller with enough
memory to handle basic cryptographic primitives that
are mandatory for the design of a security protocol.
1.1 Targeted Scenario
The targeted scenario is depicted in Figure 1. The
scenario that we target is a car sharing scheme where
a user obtains access to a vehicle from an authorized
car-sharing center and has the ability to delegate part
of his rights further. We consider that the car sharing
center can advertise the position of existing cars, i.e.,
their GPS coordinates, via a smart-phone based appli-
cation (car-sharing by smart-phone apps is a realistic
scenario and such schemes are already deployed in
practice as several third-party solutions can be found
on the web).
However, rather than relying on access via smart-
phones, which are not yet a perfect platform in terms
of security, we want to rely on microcontroller-based
RF keys which cannot be compromised by malicious
applications. That is, a user that already has a car-key
from the car-sharing center, can locate all available
cars, choose the one that is in his closest vicinity and
use it if he has the rights. Requests for the particular
vehicle can be made from the smart-phone applica-
tion, but further access is gained with the RF key.
Moreover, we want a key to express the ability to
delegate a subset of existing rights to another key wit-
Groza, B., Andreica, T. and Murvay, P-S.
Designing Wireless Automotive Keys with Rights Sharing Capabilities on the MSP430 Microcontroller.
DOI: 10.5220/0006280601730180
In Proceedings of the 3rd International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2017), pages 173-180
ISBN: 978-989-758-242-4
Copyright © 2017 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
173
User
Rights delegation
Physical car key
(with MSP430 controller)
Key issuing
Authority building
Secondary
User
Rights verification
Car-sharing center
Physical car key
(with MSP430 controller)
Car
Rights claim
Rights claim
Figure 1: Car access-rights sharing scenario.
hout involving the sharing center. Such a functiona-
lity may be useful in case when a group rents a num-
ber of cars and then at some point decides to change
cars between users or further keep a single car and
reconfigure all keys to work with the particular car.
To make access rights verifiable by neutral third par-
ties, e.g., authorities, that must be able to trace back
the key to the original user/issuer (for example in case
of misuse) we do store digitally signed access rights
on the key as soon as the user successfully claims his
rights over the first car.
1.2 Related Work
Automotive RF keys were commonly analyzed by the
research community and there are numerous results
that showed them as insecure, e.g., (Francillon et al.,
2011; Tillich and W
´
ojcik, 2012; Verdult et al., 2012;
Shoukry et al., 2013; Wetzels, 2014). Besides poor
security, these classical car keys cannot answer to the
demands of our practical scenario. They merely im-
plement simplistic challenge-response protocols that
are useless for right delegation and verification by
neutral parties. Recently, the use of smart-phone-
based technologies has been discussed (Busold et al.,
2013), (Timpner et al., 2013), (Hong et al., 2016) and
these approaches open road for complex car sharing
procedures. However, our work tries to avoid the use
of a secondary smart-device in order to save costs,
remove the need for carrying a mobile phone and
possibly increase security (as smartphones are easier
to compromise). Consequently, our protocol design
relies on distinct cryptographic constructions as we
need to adapt to a platform with low computational
capabilities.
2 PROTOCOL DESIGN
We begin by presenting the protocol design then pro-
ceed to a discussion on the choice of the crypto-
graphic primitives.
2.1 Protocol Details
The proposed protocol suite is summarized in Fi-
gure 3. Briefly, the role of the protocol components
is the following:
1. Key issuing is the procedure designated for
releasing a new car key. This happens at the car
sharing center ShCenter and we assume that it
happens in a secure environment. Therefore,
no security measures are needed on the chan-
nel. By this procedure, each key receives an
identifier ID
RFKey
ID
, a secret key shared with the
sharing center K
RFKey,ShCenter
, the access rights
rights = {raccess,time,ltime, owner:ShCenter}
and the signature from the sharing cen-
ter over the key identifier and rights
sig rights = Sig
ShCenter
ID
RFKey
,rights
. The
sharing center also stores freshness related
information such as the time of registration and
the lifetime of the user access rights, i.e., time
and ltime. While the key cannot perform time-
synchronization with a remote server and cannot
keep a secure timer (the key is dependent on
battery which may be disabled by an adversary),
we also keep this freshness related information
on the key to be verified by third parties. Since
access to the vehicle is granted with the consent
of the sharing center, it is sufficient for the sharing
center to keep a secure timer and the car can
synchronize with it.
VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems
174
2. Rights claim key-to-car is the procedure in which
the user is claiming his rights over a car. First,
the car receives the identifier of the key ID
RFKey
,
the access rights rights along with their signature
sig rights. If the signature is not authentic or the
lifetime of the key has expired, the protocol stops.
A random value rnd
RFKey
is further used in the
challenge-response protocol to ensure freshness.
The car needs to contact the sharing center for es-
tablishing a shared cryptographic key and further
sends all the information that was received from
the key along with its own fresh random value
rnd
Car
. We assume that the communication chan-
nel between the car and the sharing center is se-
cure. This is a natural hypothesis since modern
cars are equipped with 3G/4G communication and
SSL/TLS capabilities are already present on auto-
motive grade embedded devices. Designing a se-
curity protocol for the communication channel be-
tween the car and the sharing center would be out
of scope for our work. The sharing center verifies
the access rights, note that it may be that the key
acquired the rights from the sharing center or from
another key as discussed in the last procedure pro-
cedure. In the first case the sharing center has to
verify its own signature by using the public key,
in the second it has to verify a MAC (Message
Authentication Code) with the secret key that is
shared with the initial owner of the access rights.
The car receives the shared key with the RF phy-
sical key which is computed by the sharing cen-
ter as K D
rnd
RFKey
,rnd
Car
,K
RFKey,ShCenter
. Ad-
ditionally, the car receives a signature over the
granted access rights and the new owner-car pair,
i.e., RFKey,Car. Whenever the access rights ex-
pire, the car will deny access to the user. Now
the car answers in a challenge response man-
ner with a MAC computed over the participant
identities, the random values and the shared key
K
RFKey,Car
. The physical key can already compute
this shared key as it already is in possession of
K
RFKey,ShCenter
, then verify the response and ans-
wer to the challenge with a MAC over the va-
lues in reverse order, i.e., MAC
K
RFKey,Car
(ID
RFKey
,
ID
Car
,rnd
RFKey
,rnd
Car
).
3. Car access is the basic procedure in which a
user asks for a particular functionality to the
car, e.g., open doors, windows, start the en-
gine, etc. A challenge-response protocol with
the shared key is used to prove the identity
of each participant. The key simply asks the
functionalities that it wants to perform func and
the car responds with some random material
rnd
Car
. Then the key replies to the challenge
Table 1: Summary of notations.
RFKey an radio-frequency vehicle access key
Car the vehicle to which access is requested
ShCenter the vehicle sharing center
ID id associated to an RF key or car
raccess access rights to the vehicle
func functionalities (requested from vehicle)
rnd random value
Sig digital signature
K D key-derivation process
H hash function
MAC message authentication code
with a MAC computed via the secret shared
key over the identities, functionalities, its own
random material and the received random value,
i.e., MAC
K
RFKey,Car
(ID
RFKey
,ID
Car
,func, rnd
RFKey
,rnd
Car
) .
4. Rights delegation key-to-key is the procedure in
which the owner of one key, can designate a sub-
set of his rights to another key. First, the key sends
a request message with its own ID. The users have
to verbally agree and check on the display of their
keys that the key IDs are correctly set. The se-
cond key responds with the granted access rig-
hts raccess
0
and a MAC over them denoted as
sig rights = MAC
K
RFKey,ShCenter
RFKey
2
,rights
0
. Finally, the requester confirms that the requested
access rights have been received. For practical re-
asons, considering a rather classical interface for
the key, the requester can simply push the buttons
of the key corresponding to the access rights and
these are to be confirmed by the other party. While
the physical design of the key is out of scope for
our work, for clarity, we do suggest it in Figure 2.
We consider that the insecure RF channel of this
step is protected by the visual channel between
the user and the key. Since there is no secretly
shared value between the two keys, relying on
user’s feedback is the only alternative.
5. Verification by neutral third parties, e.g., authori-
ties, is not presented as a distinct protocol compo-
nent as this procedure is straight-forward from the
certification chain. The neutral third party needs
to be in possession of the public-key certificate of
the sharing center and the RF key must externa-
lize the signed access rights by the sharing center.
This can be straight-forwardly achieved.
Designing Wireless Automotive Keys with Rights Sharing Capabilities on the MSP430 Microcontroller
175
Figure 2: Suggestive depiction of the envisioned RF key
from our work.
2.2 Choice of Cryptographic Primitives
Given the limited amount of memory, i.e., 1KB of
RAM and 32KB of Flash (part of which is also needed
for other non-security related tasks) and the absence
of cryptographic hardware, symmetric key crypto-
graphy seems to be the only alternative. Symmetric
primitives can be efficiently performed on MSP430
as will be detailed in the next section. The protocol
relies only on simple hash functions and immediate
derivatives: MACs and key-derivation functions.
More flexibility can be achieved by the use of
public-key primitives, but these are for the moment
too expensive for our setup. We give here only a brief
discussion on this alternative. Asymmetric primitives,
such as the RSA or DSA, do not appear to be an op-
tion due to technical limitations. This scenario may be
interesting for reviving a cryptographic construction
that today is somewhat out-of-focus: one-time sig-
nature schemes. While several limitations exists we
briefly discuss here the possibility to use them. Such
schemes were proposed in the ’80s specifically for
highly constrained devices. While there are also se-
veral more recent proposals, none of them shows dra-
matic improvements in performance compared to the
simplest construction which is the genuine Merkle di-
gital signature scheme (Merkle, 1988). The main dra-
wback of Merkle-like signatures is that they require
significant storage for each of the signed bits.
One improvement of the Merkle scheme (see Note
11.95 from (Menezes et al., 1996)) offers a time-
memory trade-off that requires roughly 2λdk/log
2
λe
computations and a signature of size 2dk/ log
2
λe hash
outputs for signing k bits. Considering that the sig-
nature is performed over a hash (i.e., the traditio-
nal hash-then-sign paradigm) and that for short-term
use 80 bits of security are sufficient, we try to get
some rough computational estimates. For example,
by setting λ = 16 the 80 bit output to be signed leads
to a signature of 2 80/ log
2
16 80 = 2 20 80 =
3200 bits. The number of computations is 2 16
80/log
2
16 = 2 16 20 = 640 hash function compu-
tations. By using in advance some of the computa-
tional results from the following section, the compu-
tational time is under 1 second. Consequently, using
one-time signatures may be a possible alternative.
The only limitation is the one-time nature of such
signatures. Since the aforementioned signature will
require a key of 3200 bits and the controller has a
memory of 32KB, assuming that at most half of the
memory is used by the application data (a realistic
assumption), 4 such keys can be stored in memory.
By the use of Merkle trees these can be efficiently
linked to an original key, thus signing multiple times
is possible. While theoretically Merkle trees can be
extended indefinitely, this will not be possible since
storing the entire path of a signature will require too
much memory. Thus, in case of one-time signatures,
the sharing capabilities will be restricted to several
sharing operations with the controllers of our setup.
While we do rely on the simpler MAC-based proto-
col, one-time signatures may be considered for further
extensions.
3 TECHNICAL DETAILS AND
EXPERIMENTAL RESULTS
We first clarify the computational capabilities behind
MSP430 microcontrollers, then we proceed to some
details on the proposed network setup and provide re-
sults on energy consumption which are relevant for a
RF key that relies on a small battery.
3.1 Computational Capabilities Behind
MSP430
The 16-bit MSP430 family of microcontrollers fea-
tures a multitude of configurations for a wide range
of applications which require low power consump-
tion including automotive products and wireless sen-
sor networks. The computational efficiency of va-
rious cryptographic primitives running on this plat-
form was presented in several works. Some compu-
tational results on running SHA1 and SHA2 on an
MSP430 are given in (Romann and Salomon, 2014)
while (Buhrow et al., 2014) presents speed and energy
efficiency measurements for several versions of AES
and Speck. A number of papers (Hinterw
¨
alder et al.,
VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems
176
Key issuing (secure environment at car-sharing center)
1. ShCenter 99K RFKey: ID
RFKey
,K
RFKey,ShCenter
= K D(K
master
,ID
RFKey
),
rights = {raccess,time, ltime, owner:ShCenter},sig rights = Sig
ShCenter
ID
RFKey
,rights
Rights claim key-to-car (insecure environment)
1. RFKey Car: ID
RFKey
,ID
Car
,rnd
RFKey
,rights, sig rights
2. Car 99K ShCenter: ID
Car
,ID
RFKey
,rnd
RFKey
,rnd
Car
,rights, sig rights
3. ShCenter 99K Car: K
RFKey,Car
= K D
rnd
RFKey
,rnd
Car
,K
RFKey,ShCenter
,
Sig
ShCenter
ID
RFKey
,ID
Car
,rights
4. Car RFKey: ID
Car
,ID
RFKey
,rnd
Car
,MAC
K
RFKey,Car
ID
Car
,ID
RFKey
,rnd
Car
,rnd
RFKey
,
sig car rights = Sig
ShCenter
ID
RFKey
,ID
Car
,rights
5. RFKey Car: ID
RFKey
,ID
Car
,MAC
K
RFKey,Car
ID
RFKey
,ID
Car
,rnd
RFKey
,rnd
Car
Car access (insecure environment)
1. RFKey Car: ID
RFKey
,ID
Car
,func, rnd
RFKey
2. Car ShCenter: ID
Car
,ID
RFKey
,rnd
Car
3. RFKey Car: ID
RFKey
,ID
Car
,MAC
K
RFKey,Car
ID
RFKey
,ID
Car
,func, rnd
RFKey
,rnd
Car
Rights delegation key-to-key (insecure environment
a
)
1. RFKey
2
RFKey
1
: ID
RFKey
2
, rightsRequest
2. RFKey
1
RFKey
2
: ID
RFKey
1
,ID
RFKey
2
,rights
0
= {raccess
0
,time
0
,ltime
0
,owner:RFKey
1
},
sig rights = MAC
K
RFKey,ShCenter
RFKey
2
,rights
0
3. RFKey
2
RFKey
1
: ID
RFKey
2
,ID
RFKey
1
,rights
0
a
users will have to check and confirm on the visual display of the physical key that rights are shared between the correct
identities RFKey
1
and RFKey
2
Figure 3: Components of the proposed protocol suite: key issuing, rights claim, car access and rights delegation.
UART
RF
Key
MSP430
MSP430
S12
Figure 4: Expected network connectivity.
2014), (Szczechowiak et al., 2008), (Wenger and Wer-
ner, 2011) focus on the implementation and evalua-
tion of elliptic curve cryptography on MSP430 family
members proving the feasibility and limits of such ap-
proaches on this group of constraint devices.
We selected the MSP430F2274, a member of the
MSP430F2X/4X subfamily, as a candidate platform
for implementing the proposed wireless access sy-
stem. Versions of the MSP430F227 are available with
aerospace qualifications which are even more strict
than the ones in the automotive domain. With its 1KB
of RAM, 32KB of Flash and a maximum operating
frequency of 16MHz our choice fits the device cate-
gory used for automotive key applications.
In previous work (Murvay et al., 2016) we evalu-
ated the capabilities of this device for executing cryp-
Designing Wireless Automotive Keys with Rights Sharing Capabilities on the MSP430 Microcontroller
177
Table 2: Execution speed (milliseconds) on MSP430 at
16MHz of some hash functions based on previous work
(Murvay et al., 2016).
Input size Cryptographic primitive (block size)
(bytes)
MD5 SHA1 SHA2 Blake2
128 160 256 256
8 0.427 3.642 3.691 2.263
64 0.709 7.304 7.117 2.270
576 2.985 36.74 34.50 16.85
1536 7.253 91.95 85.85 44.19
4096 18.63 239.2 222.8 117.1
tographic primitives. The test results showed that
the MSP430 core can successfully handle symme-
tric cryptographic primitives as proved by measure-
ments illustrated in Table 2 which presents the execu-
tion speed for several hash functions
1
when executed
for various input sizes at an operating frequency of
16MHz. Note that here we refer to the performance
of the core since the results are only dependent on the
operating frequency and are indicative on the com-
putational capabilities of the entire MSP430 family
when the same frequency is employed.
Code size for each hash function implementation
is provided in Table 3 along with the Flash memory
occupation percentage relative to the MSP430F2274
memory size. The evaluated primitives occupy less
than 20% of the Flash memory area when no optimi-
zations are applied leaving sufficient space for imple-
menting the communication protocol and other appli-
cation features. If needed, occupied space could be
decreased by applying code size optimizations.
Table 3: Flash memory consumption of primitive imple-
mentations on MSP430 based on previous work (Murvay
et al., 2016).
MD5 SHA1 SHA2 Blake2
128 160 256 256
bytes % bytes % bytes % bytes %
6394 19.98 1338 4.18 2610 8.16 4046 12.64
Since the available RAM memory is not suf-
ficient both for program execution and for storing
cryptographic keys the Flash memory remains the
only storage alternative in the absence of an EE-
PROM. The amount of Flash memory available for
key storage is limited by the space required for storing
the actual program object code and by the foreseen
number of required re-writes of the key storage space
during the lifetime of the product. The MSP430 Flash
is guaranteed for 10.000 erase cycles. If more re-
1
MD5 and SHA1 are known to be insecure and we keep
them just as a bottom line for performance
write cycles are required this can be achieved by EE-
PROM emulation at the cost of a significantly reduced
storage space depending on the key storage structure.
Our experimental setup consists of two MSP430
Wireless Development Tools(EZ430-RF2500) made
by Texas Instruments and a NXP ZK-S12-B Kit. The
two MSP430 boards are used both for wireless com-
munication as well as for the proposed protocol im-
plementation. One of them is integrated in the car key,
while the other one is connected to the car’s BCM.
The S12 board contains a MC9S12C128 microcon-
troller and is used to stand for the BCM.
3.2 Details on Connectivity
It is relevant to point out that designing the key al-
one is not sufficient without the corresponding coun-
terpart inside the vehicle. Thus, the network design
that we consider consists in a secondary MSP430 con-
troller that is placed inside the car and communi-
cates with the BCM (Body Control Module) of the
car, a unit which is responsible to all functionali-
ties related to the body (windows, doors, trunk, etc.).
For illustration purposes we choose a board equip-
ped with a Freescale S12 core, a controller that is
used in real-world BCMs. We find that the commu-
nication between the two devices can be easily done
by using the UART (Universal Asynchronous Recei-
ver/Transmitter) interface. For our experimental se-
tup we chose to use a baud-rate of 4.8 kBaud, but the
baud-rate can be configured anywhere in a range bet-
ween 1.2 and 38.4 kBaud. This setup is suggested in
Figure 4.
A second detail on connectivity is that traditional
vehicle RF keys operate in the Ultra high frequency
(UHF) at 433 MHz. Microcontrollers from the MSP
family do posses sub-1GHz communication capabili-
ties. Our MSP430 experimental devices do commu-
nicate in the 2400-2483.5 MHz band. Using this fre-
quency band is not an issue in general as it is already
used in the automotive domain, e.g., for car alarms.
Moreover, all of the present results will hold on any
other MSP-based system core that uses a sub-1GHz
transceiver.
3.3 Energy Consumption
Figures 5 and 6 depict the energy consumption, acqui-
red by the use of an Agilent oscilloscope, as the
MSP430 microcontroller transits between several sta-
tes including a hash function computation state for
SHA1 in the former figure and SHA256 in the latter.
The voltage scale is at 50mV per division while
the time scale was set to 1s per division. The inten-
VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems
178
tion is to highlight the power consumption for crypto-
graphic functions in contrast to other functionalities.
In the first state (1), the microcontroller is in low-
power mode, then in the second state (2) the control-
ler is in normal mode and runs no tasks in the back-
ground. This is followed by a third state in which an
LED is lit (3) and the fourth state in which it executes
the corresponding hash function (4). Then in the last
state (5) several messages are sent via RF. It is easy
to see that the energy consumption for executing the
cryptographic primitive is very low compared to the
RF transmission or even lighting a 2mA rated LED.
Consequently, the cryptographic functionalities will
not have a more significant contribution to the bat-
tery depletion rate than regular tasks, e.g., reading a
button or making an LED blink. There is no noticea-
ble difference between the power consumption recor-
ded during the computations of the two hash functions
which . Moreover, executing SHA1 and SHA256 on
the MSP430 platform is very similar in terms of du-
ration (e.g. a hash execution on an 8 byte input takes
455µs for SHA1 and 461µs for SHA256). This ma-
kes the two plots look very similar although they re-
present execution cycles involving two different hash
functions.
Figure 5: Energy consumption during SHA1 computations
(4) compared with the consumption obtained for periods
of low-power mode, normal mode without tasks, turned-on
LED and during the RF transmissions.
4 CONCLUSION
Our work sets the first steps in the use of the MSP430
platform from Texas Instruments for vehicle RF keys.
So far the results prove that basic security functiona-
lities (e.g., symmetric functions, challenge-response
protocols) are straight-forward to deploy while more
advanced security functionalities (e.g., digital signa-
tures) are also within reach. The main scope of our
Figure 6: Energy consumption during SHA256 compu-
tations (4) compared with the consumption obtained for
periods of low-power mode, normal mode without tasks,
turned-on LED and during the RF transmissions.
work was to clarify some of the technical constraints
on MSP430 raised by the proposed scenario. A full-
scale implementation, security proofs for the presen-
ted schemes, as well as their redesign in case of flaws,
is a relevant subject for future work in case that the
ideas prove promising for further investigations.
ACKNOWLEDGEMENTS
This work was supported by a grant of the Romanian
National Authority for Scientific Research and Inno-
vation, CNCS-UEFISCDI, project number PN-II-RU-
TE-2014-4-1501 (2015-2017).
REFERENCES
Buhrow, B., Riemer, P., Shea, M., Gilbert, B., and Daniel,
E. (2014). Block cipher speed and energy efficiency
records on the MSP430: System design trade-offs for
16-bit embedded applications. In International Confe-
rence on Cryptology and Information Security in Latin
America, pages 104–123. Springer.
Busold, C., Taha, A., Wachsmann, C., Dmitrienko, A.,
Seudi
´
e, H., Sobhani, M., and Sadeghi, A.-R. (2013).
Smart keys for cyber-cars: Secure smartphone-based
NFC-enabled car immobilizer. In 3rd ACM Confe-
rence on Data and Application Security and Privacy,
pages 233–242. ACM.
Francillon, A., Danev, B., Capkun, S., Capkun, S., and Cap-
kun, S. (2011). Relay attacks on passive keyless entry
and start systems in modern cars. In NDSS.
Hinterw
¨
alder, G., Moradi, A., Hutter, M., Schwabe, P., and
Paar, C. (2014). Full-size high-security ECC imple-
mentation on MSP430 microcontrollers. In Interna-
tional Conference on Cryptology and Information Se-
curity in Latin America, pages 31–47. Springer.
Designing Wireless Automotive Keys with Rights Sharing Capabilities on the MSP430 Microcontroller
179
Hong, J., Shin, J., and Lee, D. (2016). Strategic manage-
ment of next-generation connected life: Focusing on
smart key and car–home connectivity. Technological
Forecasting and Social Change, 103:11–20.
Menezes, A. J., Van Oorschot, P. C., and Vanstone, S. A.
(1996). Handbook of applied cryptography. CRC
press.
Merkle, R. C. (1988). A digital signature based on a con-
ventional encryption function. In A Conference on the
Theory and Applications of Cryptographic Techniques
on Advances in Cryptology, CRYPTO ’87, pages 369–
378, London, UK. Springer-Verlag.
Murvay, P.-S., Matei, A., Solomon, C., and Groza, B.
(2016). Development of an AUTOSAR Compliant
Cryptographic Library on State-of-the-Art Automo-
tive Grade Controllers. In Proceedings of the 11th In-
ternational Conference on Availability, Reliability and
Security, ARES.
Romann, R. and Salomon, R. (2014). Salted hashes for mes-
sage authentication-proof of concept on tiny embed-
ded systems. In Intelligent Embedded Systems (IES),
2014 IEEE Symposium on, pages 42–46. IEEE.
Shoukry, Y., Martin, P., Tabuada, P., and Srivastava, M.
(2013). Non-invasive spoofing attacks for anti-lock
braking systems. In Cryptographic Hardware and
Embedded Systems-CHES 2013, pages 55–72. Sprin-
ger.
Szczechowiak, P., Oliveira, L. B., Scott, M., Collier, M.,
and Dahab, R. (2008). NanoECC: Testing the Limits
of Elliptic Curve Cryptography in Sensor Networks,
pages 305–320. Springer Berlin Heidelberg.
Tillich, S. and W
´
ojcik, M. (2012). Security analysis of an
open car immobilizer protocol stack. In Trusted Sys-
tems, pages 83–94. Springer.
Timpner, J., Sch
¨
urmann, D., and Wolf, L. (2013). Se-
cure smartphone-based registration and key deploy-
ment for vehicle-to-cloud communications. In Pro-
ceedings of the 2013 ACM Workshop on Security, Pri-
vacy and Dependability for CyberVehicles, pages 31–
36. ACM.
Verdult, R., Garcia, F. D., and Balasch, J. (2012). Gone in
360 seconds: Hijacking with hitag2. In Proceedings of
the 21st USENIX conference on Security symposium,
pages 37–37. USENIX Association.
Wenger, E. and Werner, M. (2011). Evaluating 16-bit pro-
cessors for elliptic curve cryptography. In Internatio-
nal Conference on Smart Card Research and Advan-
ced Applications, pages 166–181. Springer.
Wetzels, J. (2014). Broken keys to the kingdom: Security
and privacy aspects of rfid-based car keys. arXiv pre-
print arXiv:1405.7424.
VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems
180