Enhancing RFID Tag Resistance against Cloning Attack

Jemal Abawajy
Deakin University
School of Information Technology
Abstract In its current form, RFID system are susceptible to a
range of malevolent attacks. With the rich business intelligence
that RFID infrastructure could possibly carry, security is of
paramount importance. In this paper, we formalise various
threat models due tag cloning on the RFID system. We also
present a simple but efficient and cost effect technique that
strengthens the resistance of RFID tags to cloning attacks. Our
techniques can even strengthen tags against cloning in
environments with untrusted reading devices.
Keywords-RFID; Security; tag cloning; authentication
I. INTRODUCTION
Radio Frequency Identification (RFID) is an emerging
technology for electronic labeling of objects for the purpose of
automatically identifying, categorizing, locating and tracking
the objects. With several high-profile adoptions (e.g., Wal-
mart) and the legislative mandates for usage of RFID
technology as anti-counterfeiting technology for at-risk
pharmaceuticals in the USA [10], RFID technology has gained
significant momentum in the past few years. As a result, RFID
has emerged as practical automatic identification system in
industries as varied as high-way tolling, baggage handling,
leading clothing manufacturers and shipping companies and
cattle herding [8, 9].
With respect to the exiting identification systems such as
optical barcodes, RFID technology provides numerous
advantages that include non-line-of-sight operation, rapid
identification speed and high identification rates. For example,
in the supply chain management, RFID ensures traceability of
goods from manufacturing through sale and even beyond it.
These capabilities make RFID ideal to be used in many
application domains ranging from military to public and
commercial applications. Passive, small and inexpensive RFID
tags fulfill the needs of current supply chain applications and
are likely to serve as a future replacement of the barcodes.
Nonetheless, there are some significant challenges that
must be overcome before RFID benefits are fully realized. In
particular, despite of its enormous potentials, RFID technology
poses serious security and privacy threats both to individuals
and organizations [1, 5, 8]. Security issues have become a
barrier to the widespread adoption of RFID technology and
with the rich business intelligence that RFID infrastructure
could possibly carry, security is of paramount importance.
However, the whole RFID system security relies on the
premise that an RFID tag is harder to copy than a bar code. For
example, a passive attacker might be able to overhear the
information between a reader and some tags because the
medium is the air in the RFID system. An active attacker may
be able to send some bogus data that fakes the reader or tags to
extract information from them.
A primary security concern surrounding RFID technology
and the focus of this paper is the problem of tag cloning [5, 8,
9]. Current research in RFID is focused on privacy enhancing
technologies through access control mechanisms while security
vulnerabilities such as RFID tag cloning has received
somewhat less attention [3, 5, 8]. For example, Chris Paget has
demonstrated how one can use inexpensive off-the-shelf
components to clone large numbers of the unique electronic
identifiers used in US passport cards and next generation
drivers licenses [9]. In all these applications cloning and
impersonation of RFID tags could be financially lucrative for
occasional hackers or professional criminals, and severely
damaging for the licit companies' revenues and reputation [5].
Therefore, as tag cloning is one of the serious security
problems facing RFID infrastructure, techniques for
strengthening RFID systems against tag cloning attacks is of
paramount importance.
We propose a simple but effective new technique to tackle
the passive RFID-tag cloning problem. The proposed approach
is extremely efficient lightweight tag authentication protocol
and offers an adequate security level for certain applications
and can be implemented even in the most limited low-cost
RFID tags. In addition, the proposed scheme is compliant with
the EPC Class-1 Generation-2 UHF-RFID standard which is
the major RFID standards body known as EPCglobal has
recently ratified [2].
The rest of the paper is organized as follows. In Section 2,
the background to the RFID is presented. The system model,
threat models and related work are discussed. The proposed tag
cloning resistance algorithm and the security analysis of the
proposed algorithm are discussed in Section 3 and Section 4
respectively. The conclusion and future direction is discussed
in Section 5.
II. BACKGROUND
In this section, we specify the security properties that we
want our protocol to achieve. We begin by describing the
system model as well as the assumptions regarding the readers
and tags being used. We then discuss various threat models due
tag cloning on the RFID system.
A. System Model
As shown in Fig. 1, a typical deployment of an RFID
system involves three types of entities: tags, readers and back-
end servers. An RFID reader can be a PDA, a mobile phone or
2009 Third International Conference on Network and System Security
978-0-7695-3838-9/09 $26.00 2009 IEEE
DOI 10.1109/NSS.2009.101
18
2009 Third International Conference on Network and System Security
978-0-7695-3838-9/09 $26.00 2009 IEEE
DOI 10.1109/NSS.2009.101
18
2009 Third International Conference on Network and System Security
978-0-7695-3838-9/09 $26.00 2009 IEEE
DOI 10.1109/NSS.2009.101
18
any kind of devices capable of querying object identity stored
in a RFID tag.
The tags are attached to, or embedded in, objects to be
identified. Each tag in the system has a unique identifier. RFID
tags can store basic information (e.g., an address) as well as
more complex information as well as more complex
information such as a the repair and service history of an
aircraft part [18]. In general, RFID tags are classified based on
their power supply source, as passive, semi-passive, and active.
Passive tags have no internal power source while semi-passive
and active tags have on-board power source. Cost wise, passive
tags are the cheapest and expected to soon be comparable to
that of barcodes. This makes passive tags quite attractive to
businesses. An example of passive tags is an EPC (Electronic
Product Code) tag [2]. As low-cost RFID tags are expected to
be in wide use, we will specifically focus on them in this paper.
Reader
Tag
Backend
Server
Secured Wired Channel
Insecure RF
Channel
Fig. 1: An RFID system architecture
The tags and the reader communicate using a radio
frequency link, while the reader is connected to the back-end
server through a secure wired link. Note that securing the
communication link between the readers and the back-end
database could be trivially accomplished by protocols such as
SSL. The back-end servers are trusted entities that maintain an
enterprise database containing the information needed to
identify tags, including their identification numbers. It is
assumed that the servers are physically secure and not
attackable. The enterprise database hosted on the backend
server is secured by authentication and authorization tools and
could only be accessed by a reader after full authentication.
Due to the RFID industrys desire to trade-off hardware
functionality in order to manufacture passive tags that cost less
than 5, security has received somewhat less attention. This has
resulted in various vulnerabilities that can be easily exploited to
compromise RFID-enabled system infrastructures. In the
following subsection, we will focus on threats due to tag
cloning attacks.
B. Threat Model
RFID systems are susceptible to a broad array of
malevolent attacks ranging from passive eavesdropping to
active interference. RFID has the vulnerabilities of any wireless
system, where passive and active attacks can be performed
easily. Also, the limitations on cost, size and power
consumption for low cost passive tags result in low security
features. The whole security relies on the premise that an RFID
tag is harder to copy than a bar code. However, passive tags are
easily forgeable and creating a fake (i.e., clone) tags from an
original tag is already a simple exercise.
To cloning a tag, the tag id is essentially all the attackers
need. Juels [1] shows that EPC tags are vulnerable to
elementary cloning and counterfeiting attacks yet RFID tag
cloning has received somewhat less attention [3]. Tag cloning
attacks is exacerbated by the fact that the tags can be
manipulated by rogue readers. As an RFID readers are easily
available and even being included in mobile phones [16],
tracking a tag bearer is quite possible today. Hence it is
possible for an adversary to read the RFID tag and correlate its
time and place to learn more about the tag. Therefore, it is quite
clear that new countermeasure approaches need to be
developed to thwart the potential threats RFID systems face.
There are several issues that facilitate the tag cloning
attack, including eavesdropping threat and skimming threat.
Once tag identification is captured, an attacker can duplicate
genuine tags and use the cloned tag for a variety of malicious
purposes. For example, it can be used for counterfeiting where
an attacker can replace tags for expensive items with tags for
cheaper ones. It can also be used as a warmware. It doesn't
take much malicious code to do things like an SQL injection
in RFID, which will create a chain reaction of corrupted data
that flows into the central information resource. As
demonstrated in [6], a good RFID tag could be replaced with a
virus coded tag that could do its dastardly deeds in RFID
system. RFID tag cloning in turn could lead to other attacks
such as denial-of-service (DoS) attack. DoS occur when an
RFID system cannot function properly to provide normal
services to valid users; it is a common threat to Internet server
systems.
One major problem with RFID infrastructure is that there is
no mechanism to distinguish between real and counterfeit
components in the system. Specifically, it is impossible for an
RFID reader to distinguish an authentic tag from a cloned one.
The main security problem that we address in this paper can be
stated as follows: How can a reader tell that a certain RFID tag
is really the one that it intended to talk to? Specifically, the
question is how to prevent the cloned tags from impersonating
legitimate tags. Tag cloning attacks is exacerbated by the fact
that the tags can be manipulated by rogue readers. As an RFID
readers are easily available and even being included in mobile
phones [16], tracking a tag bearer is quite possible today.
Hence it is possible for an adversary to read the RFID tag and
correlate its time and place to learn more about the tag.
Therefore, it is quite clear that new countermeasure approaches
need to be developed to thwart the potential threats RFID
systems face.
C. Related Work
Given low-cast RFID tags expected wide-spread use in
various applications, cost effective techniques to strengthen
passive tags against tag cloning attack is paramount.
Strengthening passive tags against cloning requires making the
RFID system resistant to several attacks including skimming
attacks, eavesdropping and active attacks.
19 19 19
A natural solution to the security vulnerability stated
before is to employ cryptographic protocol in RFID
system. To this end, several implicit tag authentication
algorithms have been discussed in the literature [1, 4, 5].
In hash-lock protocol [5] and its variants, tags are
expected to respond to readers query by generating a
hashed value of its identifier. The reader will then use the
back-end server to authenticate the tag. In addition to the
fact that hash function implementation is still beyond
current capability of low-cost RFID tag, this approach has
many problems including synchronization problem,
scalability problem, spoofing and eavesdrop attacks.
The typical EPCglobal EPC Class-1 Gen-2 standard
authentication mechanism is fallible as it was shown that
after simple computations an attacker can acquire the
access and kill passwords with high probability [11]. Juels
[1] proposal for solving cloning attack is vulnerable to
eavesdropping and active attacks, etc. Karthikeyan and
Nesterenko [12] proposed a tag identification and reader
authentication protocol based on simple XOR and matrix
operation. This approach could end up updating the key
wrongly leading to the situation where the legitimate
reader and the tag will not be able to authenticate each
other anymore. Moreover, the protocol is also vulnerable
to replay attacks and privacy location is not guaranteed.
In the minimalist cryptography approach [4], a tag
stores a set of pre-prepared pseudonyms. When a tag is
queried by a reader, it responds with the next metaID in
the list. The main flaw of this scheme is that an attacker
can send a few queries to obtain all the metaIDs on a tag
and use these metaIDs to track this tag until the tag meets
a legitimate reader again. The solution proposed in [1] to
modify EPC Class-1 Generation-2 tags, to test the validity
of the kill PIN, is operable only with EPC tags. The
solution does not protect against eavesdropping and
cloning of a tag. Also the solution is only temporary.
Passive RFID tags are designed to offer battery-free
operation and low cost. Thus, they have extremely
limited capabilities often being able to holdand in fewer
cases protectonly a simple entity identifier and
potentially a very limited amount of information. Also,
the logic gates of the current passive tags are about 500
5000 and thus RFID tags require lightweight encryption
and security protocols. This limitation coupled with
limited computing resources and cost, existing encryption
and security protocols are not practical for use on limited
resource tags. Therefore, an authentication protocol must
be designed taking into consideration the limited
computing capacity and low storage space of low price
tags. Therefore, none of the existing proposals are an ideal
solution to the problem we stated. Also, given the current
and expected low-cost RFID tag capabilities, most of the
existing proposals could not be practical as they assume
beyond the capabilities of passive tags.
Our scheme is based on partial information sharing
between the entities in the system. The proposed protocol
protects against tag cloning by protecting the identity of
tags. Additionally, our protocol is very simple and it can
be implemented easily with the use of standard
cryptographic hash functions.
III. TAG CLONING RESISTANCE ALGORITHM
In this section, we discuss the proposed tag cloning
resistance protocol. Before describing our protocol in detail,
we give the definitions of notations that we use in the
description of our protocol. We assume that there are Y
readers in the system:
( )
Y
R , , R R ...
1
=
(1)
Also, we assume that there are X passive tags in the system:
( )
X
T , , T T ...
1
=
(2)
Each tag
T T
i
e
has an electronic product code ( )
EPC
T . As
passive tags are resource limited devices that only have less
than 1K logical gates for security functions, we assume that a
tag is able to perform rudimentary functions such as string
comparison, bitwise XOR ( ) and pseudo-random number
generation (PRNG),
{ } { }
l
f 1 , 0 1 , 0 :
*

(3)
These assumptions are justified by current features
implemented in many RFIDs tags [2]. For a PRNG that
generates a 16-bit pseudo-random number, it is shown that,
among a number of 10,000 tags, the chance that any two tags
simultaneously generate the same 16-bit pseudo-random
number is less than 0.1% and the probability of that guessing
the next pseudo-random number generated by a tag is less than
0:025% under the assumption that all previous outputs are
known to an attacker [2].
We assume that the readers and the server are able to
perform, in addition to those operations performed by the tag,
standard cryptographic protocols and one-way hash function
operation given as follows:
{ } { }
l
h 1 , 0 1 , 0 :
*

(4)
The RFID reader and the server will run the same hash
function and the same random number generator with the
same seed S.
The proposed algorithm has two main phases: registration
phase and communication phase, each of which is described
below.
A. Pre-deployment phase
Tags and readers must register with the database server
separately before deployment. We assume that the registration
process is performed under a secure environment.
Database server
The server generates and maintains in its database over all
known tags the following list of tuple:
( )
i ID EPC VID
K R , T , T D , (5)
where
20 20 20
-
EPC
T and
VID
T denote the actual an virtual unique
identifiers of tag T T
i
e respectively;
-
ID
R
and
VID
R
denote the actual and virtual unique
identifiers of reader R R
i
e ; and
-
i
K denotes the currently valid key for tag T T
i
e and
R R
i
e .
We assume that the backend server is highly trustful. The
key
i
K is generated uniformly at random, and held as a shared
secret between the tag and the server. Also, we assume that
during the manufacturing time, the manufacturer setup a tag by
assigning
EPC
T and other parameters. Similarly, the reader will
have a unique identifier assigned by the manufacturer. Note
that both of these values could also be done by the server prior
to the tag and reader deployment.
Button Si,RID, RVID, Ki
Button TVID, TEPC, Ki
..
.
TVID, TAID, RID, Ki
.
.
TEPC
Database
Tag
Reader
RVID,Si, Ki
RID
TVID, Ki
Fig. 2: Pre-deployment registration
Fig 2 shows the steps undertaken at pre-deployment
registration point. Upon registration, both the tag and the reader
get a set of core parameters needed to distinguish a family from
non-family system component from the server and these
parameters are then stored in the database.
Reader Registration
Each reader R R
i
e will maintain the following tuple:
( )
i VID ID i i
K R R S R , , , (8)
Upon registration with the server, the reader will receive the
following tuple from the server:
( )
i VID i i
K R , S R , (9)
where
VID
R is the virtual unique identifier of the reader and it
is computed as follows:
( ) ( ) ( )
i R ID L ID VID
S R D h R (10)
where ( )
L ID
D is the left half of the database id and ( )
R ID
R is
the right half of a registered reader id. It is assumed that
both ( )
L ID
D and ( )
R ID
R are two large numbers and unique in
the system. For example, let us say ( ) 1010 =
L ID
D and
( ) . 0100 =
R ID
R Then, ( ). 10100100 h T
VID

The parameter
i
S is the current session nonce, which is
renewed each time the server and the reader communicate and
generated as follows:
( )
i i
S f S
+1
(10)
Tag Registration
Each tag T T
i
e will maintain the following tuple:
( )
i EPC VID i
K T , T R , (10)
At registration point, the server assigns each tag the following
tuple:
( )
i VID i
K , T T (11)
The parameter
VID
T
denotes the tag virtual id, which is
computed as follows:
( ) ( ) ( )
R ID L ID VID
R D h T (7)
where ( )
L ID
D is the left half of the database id and ( )
R ID
R is
the right half of a registered reader id..
B. Tag Authentication Algorithm
Fig. 3 shows the basic steps interactions between the RFID
system components after deployment. The two principal parties
involved in this protocol are readers and tags. Due to the fact
that the tags used are passive, the communication must be
initiated by the readers.
Algorithm TagAuth
Begin
1. ( ) ( ) ( ) S f T T T:T
v L a i

2.
i
R:T T
3. ( ) { R R R:
k R k
e
s
a. ( ) ( )
R k L a i s
R T T K .
b. ( ) ( )
v i s
T T K h C . ==
c. ( ) break True C if ==
}
4. Return C
End TagAuth
Fig. 3: Tag authentication algorithm
How can a reader tell that a certain RFID tag is really the
one that it intended to talk to? The answer to this question
requires the existence of a tag authentication protocol that is
capable of effectively preventing the tag cloning attack
mentioned above. In this section, we discuss the proposed tag
authentication protocol.
We assume that there are x tags,
{ }
x
T T T ,...,
1
=
and y
readers,
{ }
y
R R R ,...,
1
=
in the system. Each tag T T
i
e
maintains a ( )
i a v
K , T , T tuple where
a
T denotes the actual
21 21 21
unique identifier of tag
T T
i
e
whereas
v
T
denotes the tag
virtual id and computed from the left half of
a
T
and the right
half of
R R
i
e
as follows:
( ) ( ) ( )
R i L a v
R T h T
(1)
It is assumed that both ( )
L a
T and ( )
R i
R two large numbers
and unique in the system. Similarly, the server maintains a list
of
( )
i i a v
K R , T , T ,
tuples in its database over all known tags.
i
K denotes the currently valid key for tag
a
T . We assume that
i
K is generated uniformly at random, and held as a shared
secret between the tag and the server.
As tags are resource limited devices that only have less than
1K logical gates for security functions, we assume that a tag is
able to perform rudimentary functions such as string
comparison, bitwise XOR ( ) and random number generation
(PRNG), { } { }
l
f 1 , 0 1 , 0 :
*
. This is justified by current
features implemented in many RFIDs tags [2].
In contrast, the readers and the server are able to perform, in
addition to those performed by the tag, one-way hash function
operation { } { }
l
h 1 , 0 1 , 0 :
*
and standard cryptographic
protocols. The reader and the server will run the same hash
function and the same random number generator with the same
seed S.
The proposed authentication protocol called TagAuth is
presented in Fig. 3. Following [1], B A indicates a data
flow from entity A to entity B, while A: indicates an operation
performed locally by A. Due to the fact that most low-cost tags
are passive, the communication must be initiated by readers.
Thus, TagAuth is executed by a RFID reader R attempts to
authenticate a tag T.
Specifically, when interrogated by a reader, a tag generates
i
T and sends it to the reader as follows (steps 1-2):
( ) ( ) ( ) S f T T T
v L a i
(2)
In Eq. 2, the parameters ( )
L a
T and
v
T denote the left half bits
of the tag identifier and the virtual id of the tag. The function
( ) S f generates fresh random number.
When the reader receives
i
T , it goes into tag authentication
process. The reader first creates a secrete key ( )
s
K from
i
T
and the reader identifier as follows:
( ) ( )
R k L a i s
R T T K . (3)
where ( )
L a i
T T . is the left half bits of the tag id while ( )
R i
R is
the right half bits of reader k. It then creates a hash of
s
K and
compares it with
v i
T T . . If a match is found, the reader accepts
the tag as authentic. Otherwise, it rejects it.
Our protocol is practical and useful for two reasons. First, it
requires only an extremely small amount of computation;
therefore, it has the capacity to be implemented within passive
RFID tags, unlike the cryptography intense protocols in prior
work. Second, our protocol deals with both privacy and
authentication. This also decreases the overall cost production;
but more importantly, it eliminates the need for any extra
security devices.
IV. SECURITY ANALYSIS
Most RFID systems are inherently insecure. Our scheme is
based on partial information share between the entities in the
system. The proposed protocol enforces user privacy and
protects against tag cloning. Additionally, our protocol is very
simple and it can be implemented easily with the use of
standard cryptographic hash functions.
It is obvious that our authentication scheme satisfies many
of the security requirements we mentioned in the introduction,
in particular protection against tracing and tag cloning. Since
the tag response changes in a secure manner after every read
query then attempts like eavesdropping, spoofing, replaying
messages, etc. cannot compromise the security of the scheme.
Thus, our scheme does not suffer from the obvious security
disadvantage of tag traceability.
A tag that does not carry a valid identifier will not achieve
successful authentication in this protocol. Thus an adversary
cannot successfully clone a tag without knowledge of a valid
a
T obtained, for example, via skimming. However, the tag
never sends its identifier thus, it is quite difficult to clone a tag.
Listening to messages exchanged in a particular session
reveals no information. Furthermore, with every session a new
nonce is generated guaranteeing the freshness of messages.
Therefore, if the tag which changes its response at every
session using a random nonce opens a new session for a fake
request message, the ongoing legal authentication session will
fail by this illegal authentication trial.
Furthermore, since a tag response is randomized in every
session, our scheme prevents an adversary from performing a
spoofing attack. The protocol makes it hard to mount a replay
attack as the tag generates a fresh random number r in its
response. It is relatively hard for replay attacks when fresh
random values are generated and used during each new
authentication process.
V. CONCLUSIONS
RFID technology presents a number of advantages, but also
opens a huge number of security problems that need to be
addressed before its successful deployment. In this paper, we
have designed a tag authentication protocol and shown that the
proposed protocol do handle the identified threats (i.e., tag
cloning). Since RFID tag has only small computing power, our
approach to move tags computation to the reader and database
server is reasonable. Also, the proposed authentication
algorithm does not have any database synchronization problem,
which makes it ideal for implementation on low-cost RFID
tags. We discussed security aspects of our scheme and
compared its efficiency with previous ones in terms of the time
and the space complexity with the communication cost
involved in identifying one single tag.
22 22 22
ACKNOWLEDGMENT (HEADING 5)
I appreciate the help of Maliha Omer without whom this
paper would not have been possible.
REFERENCES
[1]. A. Juels, Security analysis: Strengthening EPC tags against
cloning , Proceedings of the 4th ACM workshop on Wireless
security (WiSe '05), 2005.
[2]. EPCTM, 2008 EPC (Electronic Product Code) Class1
Generation2 standard by EPCglobal, 2008. access available on
15 April 2009.
[3]. Daniel V. Bailey, Ari Juels: Shoehorning Security into the EPC
Tag Standard, Security of Communication Networks (SCN
2006), pp: 303-320, 2006.
[4]. A. Juels, Minimalist Cryptography for Low-Cost RFID Tags
(Extended Abstract), Volume 3352/2005, Security in
Communication Networks, Springer-Verlag,, Page(s): 149-164,
2005.
[5]. Mikko Lehtonen, Daniel Ostojic, Alexander Ilic and Florian
Michahelles. Securing RFID Systems by Detecting Tag Cloning,
Proceedings of 7th International Conference on Pervasive
Computing, 291-308,2009.
[6]. M. R. Rieback, B. Crispo, Tanenbaum, A. S., Is Your Cat
Infected with a Computer Virus?, IEEE Conference on
Pervasive Computing and Communications (PerCom 2006), pp.
169-179, 2006.
[7]. Nokia 5140 RFID Reader. Available at:
http://www.mobilemag.com/content/100/104/C2607.
[8]. K. Koscher, A. Juels, T. Kohno, and V. Brajkovic. EPC RFID
Tags in Security Applications: Passport Cards, Enhanced
Drivers Licenses, and Beyond. 2008.
[9]. Passport RFIDs cloned wholesale by $250 eBay auction spree,
http://www.theregister.co.uk/2009/02/02/low_cost_rfid_cloner/
[10]. M. C. O'Connor. Pfizer Using RFID to Fight Fake Viagra.
RFID Journal, January 6th, 2006.
[11]. Pedro Peris-Lopez, Tieyan Li, Julio C. Hernandez-Castro, Juan
M.E. Tapiador, Practical attacks on a mutual authentication
scheme under the EPC Class-1 Generation-2 standard Computer
Communications, Available online 5 April 2009.
[12]. S. Karthikeyan and M. Nesterenko, RFID security without
extensive cryptography, Proc. of SASN '05, ACM (2005), pp.
6367.
[13]. Hung-Yu Chien and Che-Hao Chen, Mutual authentication
protocol for RFID conforming to EPC class-1 generation-2
standards, Computer Standards & Interfaces vol. 29, Elsevier
Science Publishers (February 2007), pp. 254259 (2).
[14]. George Roussos and Vassilis Kostakos, rfid in pervasive
computing: State-of-the-art and outlook, Pervasive and Mobile
Computing, 5 (1), pp: 110-131, 2009.
[15]. E.W. Schuster, S.J. Allen and D.L. Brock, Global rfid: The
Value of the EPCglobal Network for Supply Chain
Management, Springer, 2007.
[16]. Nokia 5140 RFID Reader. Available at:
http://www.mobilemag.com/content/100/104/C2607.
[17]. F. Thiesse, E. Fleisch and M. Dierkes, LotTrack: rfid-based
process control in the semiconductor industry, IEEE Perv.
Comp. 5 (1) (2006), pp. 4753.
[18]. C. Legner and F. Thiesse, rfid-based maintenance at Frankfurt
airport, IEEE Perv. Comp. 5 (1), pp. 3439, 2006.
[19]. A.X. Liu, L.A. Bailey, PAP: A privacy and authentication
protocol for passive RFID tags, Comput. Commun. (2009),
doi:10.1016/j.comcom.2009.03.006
[20]. A. Juels, RFID security and privacy: a research survey, IEEE
Journals on elected Areas in Communications 24 (2) (2006)
381394.
[21]. Barber, E. Tsibertzopoulos, B. Hamilton, An analysis of using
EPCglobal class-1 generation-2 RFID technology for wireless
asset management, in: Military Communications Conference,
vol. 1, 2005, pp. 245251.
23 23 23