If Objects Could Talk: Semantic-enhanced
Radio-frequency IDentification
Michele Ruta, Tommaso Di Noia, Eugenio Di Sciascio and Floriano Scioscia
Politecnico di Bari
via Re David 200, I-70125
Bari, Italy
Abstract. We propose to extend basic RFID usage by storing semantically an-
notated data within RFID tags memory, so that objects may actually “describe
themselves” in a variety of scenarios. In particular here we exploit our approach
to carry out an advanced discovery process using annotations stored in RFIDs.
A –fully backward compatible– modification to the original RFID data exchange
protocol is presented, integrated in a semantic-enabled Bluetooth resource dis-
covery framework. Motivations and benefits of the approach are outlined in a
u–commerce context.
1 Introduction
Radio-Frequency IDentification (RFID) is an emerging technology interconnecting via
radio two main components: (1) a transponder carrying data (tag) located on the ob-
ject to be identified; (2) an interrogator (reader) able to receive the transmitted data.
Traditional RFID applications have been focused on supply chain management and as-
set tracking [1]. Nowadays tags with higher memory capacity and on-board sensors
disclose new scenarios and enable further applications. We present a semantic-based
environment where tagged objects become resources exposing to a reader not simply
an identification code but a semantically annotated description. It may enable objects
equipped with RFID tags describe themselves in a variety of scenarios e.g., during sup-
ply chain management, shipment, storing, sell and post-sell, without depending on a
centralized database. In particular in this paper we focus on an innovative, semantic-
based discovery mechanism. Current identification methods are largely inefficient for
advanced applications. Hence we adapt both ideas and technologies from the Semantic
Web in a u–marketplace
1
, where objects endowed with RFID tags have been dipped
into Bluetooth mobile ad-hoc networks. Building on previous works that enhanced the
basic discovery features of Bluetooth with semantic-based discovery capabilities [2],
in this paper we propose an extension of EPCglobal specifications for RFID tag data
1
Here we intend a u–marketplace as an ad-hoc environment where mobile peer users –both
buyers and sellers– can submit their advertisements, browse through available ads and be as-
sisted in finding the best available counterparts to meet their needs and initiate a commercial
transaction.
Ruta M., Di Noia T., Di Sciascio E. and Scioscia F. (2007).
If Objects Could Talk: Semantic-enhanced Radio-frequency IDentification.
In Proceedings of the 1st International Workshop on RFID Technology - Concepts, Applications, Challenges, pages 25-34
DOI: 10.5220/0002431300250034
Copyright
c
 SciTePress

standards, providing semantic-based value-added services, and present its deployment
in an advanced u–commerce setting.
The remaining of the paper is structured as follows. In the next section we comment
on related work. Section 3 outlines the proposed framework. In Section 4 we sketch
proposed enhancements. A case study clarifies our approach in Section 5. Section 6
closes the paper.
2 Related Work
U–commerce is based on the ubiquity, universality, uniqueness, and unison network
characteristics as pointed out in [3]. There, authors forecast that when network devices
and infrastructures will become more ubiquitous, they will be powerful and useful mar-
keting instruments. Nowadays RFID is envisioned as one of the most promising tech-
nologies to build u–commerce infrastructures. Nevertheless, currently it is trivially used
as a link between physical objects and a “virtual counterpart” in the digital world [4].
Most of literature focuses on innovative RFID applications.
In [5] a pervasive architecture for tracking mobile objects in real-time is presented.
Proposed solutions are aimed at supply chain and B2B transaction management. A
global and persistent IT infrastructure is necessary in order to interface RFID systems
through the Internet. These requirements make the approach less suitable for B2C and
C2C scenarios especially in mobile ad-hoc contexts.
R
¨
omer et al. [4] present a Java and a .NET frameworks for ubiquitous comput-
ing applications using smart identification technologies. Core design abstractions such
as object location, neighborhood, composition, history and context enhance flexibility.
Nevertheless, as admitted by the authors, scalability issues are present. They may be
related to a virtual counterpart approach, which seems to be unsuitable to really mo-
bile applications. A further limitation is in use of Jini and UDDI as discovery protocols
because they only support exact service matches. Semantics of object properties and
capabilities is not explicit, but it is encapsulated in either Java classes or Web Services.
Several efforts have been put on exploiting RF technologies to enhance Human-
Computer Interaction (HCI) e.g., in wearable computer architectures. Hum [6] early
introduced an OSI-like protocol stack he called Fabric Area Network (FAN), support-
ing a dynamic data routing between RFID tags deployed on garments and a single
wearable base station. Schmidt et al. focused on implicit HCI, taking user activity in
the real world as input to computers. In [7] the authors introduce a wearable RFID so-
lution enabling operations over an information system simply by picking up or using
an operation-related tagged object. The proposed system has been also integrated with
SAP R/3 in a case study. Since no semantic information are associated to tags, a virtual
counterpart is anyway needed.
In [8], interaction patterns between users endowed with GSM phones and common
objects are investigated. Objects are endowed with active RFID transponders equipped
with on-board sensors and Bluetooth connectivity. An infrastructure enabling a hybrid
implicit-explicit HCI model is implemented. Interaction is mediated by typical mobile
phone patterns (called interaction stubs), such as SMS templates and reminder alarms.
26

Proposed approach alters normal relationships between people and things. The need for
a costly communication link such as GSM is an open issue.
In [9] a support system aimed at enhancing information exchange in conferences is
presented. A location and time aware middleware tracks participants while entering or
exiting meetings by means of RFID badges and a reader deployed in each room. Access
to a shared chat session and to a remote file system folder is then granted dynamically, as
long as users stay in a room to attend an event. Context-awareness relies on preliminary
explicit profiling of both users and events of interest. Nevertheless, user experience is
enhanced without modifying people habits or their interactions with the environment.
3 Framework: Motivation and Explanation
Novel ubiquitous paradigms call for efficient and effective discovery of resources and
services available in an area. To discover resources and services, basic mechanisms
currently used are often ineffective, as they are usually based on unique IDs to identify
items and simple string matching for discovery. More powerful discovery infrastruc-
tures are desirable, able to cope with rich descriptions associated to resources in ad-
vanced scenarios. To this aim, in [2] a semantic-enhanced Bluetooth discovery protocol
was introduced, which allows a semantic-enabled discovery mechanism of resources
and services. Unused classes of service identifiers of the original standard (UUIDs)
were exploited as ontology markers, naming these values OUUIDs (Ontology Univer-
sally Unique IDentifiers). Currently, RFID technology is trivially used to unambigu-
ously identify physical objects and to retrieve related attributes by way of a fixed server.
Nevertheless, we believe there is room for more advanced and useful applications of
RFIDs extended with structured descriptions, so that a good equipped with an RFID
can semantically describe itself along its life-cycle. We conceived a unified frame-
work where an RFID-based infrastructure and advanced Bluetooth service discovery
are virtually “interconnected” at the application layer permitting advanced services in
u–environments. In particular, here, we present an extension of EPCglobal standard
[10] allowing a semantic-based object discovery in a u–marketplace framework. Pro-
tocols to read/write tags have been preserved maintaining original code-based access
(so keeping a backward compatibility with legacy applications practically without any
modification). A good can be easily and thoroughly described by means of a semantic
annotated description stored within the tag it is associated with. Hence a RFID reader,
scanning characteristics of a selected product, enables a further discovery phase iden-
tifying resources similar to the chosen one or to be combined with it. Via the semantic
based Bluetooth SDP and exploiting non standard inference services devised in [11]
best matching resources of the marketplace will be discovered and returned to the user.
Such an approach provides several benefits. Information about a product is struc-
tured and complete; it accurately follows the product history within the supply chain
being progressively built or updated during the good life cycle. This improves trace-
ability of production and distribution, facilitates sales and post-sale services thanks to
an advanced and selective discovery infrastructure. Traditional approaches, differently
from the one we propose here, do not consider items potentially matching with a user
request as well as they do not contemplate the possibility of suggesting combinations
27

SHOP
hotspot
middleware
RFID tag
reader
Fig.1. Allowed interaction sequence diagram.
of items in order to satisfy a user need. The above logical framework will be illustrated
and motivated in a virtual consumer electronics store case study. Figure 1 shows the
main elements of our prototype and a high-level view of the interaction pattern enabled
by the proposed approach. In our case study, a “smart shopping cart” is equipped with a
sensor and a tablet computer, which integrates a RFID reader and Bluetooth connectiv-
ity. When customer picks up a product, the system assists her in discovering additional
items, either similar or to be combined with the selected one. A zone resource provider
(hotspot) keeps track of resources within the marketplace. It interacts with the shopping
cart through semantic-enhanced Bluetooth, replying to its requests at SDP layer.
In the rest of the paper we will assume the reader be familiar with basic syntax
and semantics of Semantic Web languages –in particular OWL [12] and DIG [13]–
and of Description Logics (DLs) [14] which is the formal language we adopt. Here we
formalize examples by adopting DL syntax for the sake of compactness.
4 Semantic-enhanced EPC Standard
We refer to the EPCglobal standard for class I Generation-2 UHF RFID tags [10]. Mem-
ory is organized in four logical banks [15]: (1) Reserved, storing –if present– kill and
access passwords; (2) Electronic Product Code (EPC); (3) Tag IDentification (TID),
storing tag manufacturer and model identification codes; (4) User, storing –if present–
data defined by the user application. We exploit two bits in the EPC tag memory area
now reserved for future purposes. The first one –at 15
h
address– is exploited to indicate
whether the tag has a user memory (bit set) or not (bit reset). The second one –at 16
h
address– is set to mark semantic enabled tags. In this way, a Select command
2
(see Table
2
By means of it a reader imposes to each tag in range to perform a comparison between a bit
mask (Mask parameter) and the content of a tag memory area identified by the triple MemBank
(one of the four tag memory banks), Pointer (initial address within the specified bank) and
Length. Then the tag will set/reset one of its status flags according both to the comparison
outcome (match/no-match) and to Target and Action parameters. Target indicates the flag to be
28

Table 1. SELECT command able to detect only semantic enabled tags.
PARAMETER Target Action MemBank Pointer Length Mask
VALUE 100
2
000
2
01
2
00010101
2
00000010
2
11
2
DESCRIPTION SL flag set in case of match, EPC memory initial address number of bit bit mask
reset otherwise bank to compare
Table 2. READ command able to extract OUUID from the TID memory bank.
PARAMETER MemBank WordPtr WordCount
VALUE 10
2
000000010
2
00001000
2
DESCRIPTION TID memory bank starting address read up to 8 words (128 bit)
1) allows a reader to easily distinguish semantic based tags. MemBank value identifies
the EPC memory bank (bank 01
2
); Pointer is the starting address of the bit pair (since
00010101
2
= 15
h
); Length is 2. Mask is 11
2
since semantic-enabled tags are identified
by having both bits asserted. Target and Action parameters have the effect to assert the
SL tag status flag only for semantic-enabled tags and deassert it for remaining ones. The
following inventory step will skip tags having SL flag deasserted, thus allowing a reader
to identify only semantic-enabled tags (protocol commands belonging to the inventory
step have not been described, because they are used in the standard fashion).
The EPC standard for UHF - class I tags impose the content of TID memory up
to 1F
h
bit is fixed. As said above, optional information could be stored in additional
TID memory; it generally consists in serial numbers or manufacturer data. We use the
TID memory area starting from 100000
2
address. There we store the identifier of the
ontology w.r.t. the description contained within the tag is expressed. Recall that each
semantically annotated resource description is referred to a specific ontology which is
universally labeled by means of the OUUID identifier. That is the OUUID in the TID
bank marks the reference ontology w.r.t. is expressed the description of the good. In or-
der to make RFID systems compliant with the ontology support system proposed in [2],
we define a bidirectional correspondence among OUUIDs stored in RFID transponders
and those managed by Bluetooth devices. To retrieve the OUUID value stored within a
tag, a reader will exploit a Read command with parameters as in Table 2.
Together with the semantically annotated object description expressed in DIG, within
the user memory bank will be stored contextual parameters (i.e., numerical values
whose meaning depends on the specific application). Due to verbosity of DIG format
and limitations of tag memory, the use of a compression algorithm is needed. For the
sake of conciseness, here we omit characteristics of the encoding tool. To store a se-
mantically annotated description containing up to 50 concept and roles we estimate a
memory occupancy not exceeding 8 kbit. A reader can perform extraction and insertion
of a description on a tag, by means of one or more Read or Write commands respec-
tively. Both commands are compliant with the RFID air interface protocol. In Table 3,
parameters of the Read command
3
for extracting a compressed description are reported.
updated, while Action tells to the tag how to update it (i.e., whether to assert, deassert or leave
the flag unmodified).
3
It allows to read one or more 16-bit memory words from one of the four tag memory banks.
MemBank parameter identifies the memory bank to be read (as in Select command). WordPtr
29

Table 3. READ command able to extract the semantic annotations from the user memory bank.
PARAMETER MemBank WordPtr WordCount
VALUE 11
2
000000000
2
00000000
2
DESCRIPTION user memory bank starting address read up to the end
Table 4. Mapping of product categories to values of contextual resource parameter.
Value 1 2 3 4 5
Product category telephony computers photography audio and video hobbies
4.1 Semantic-based Object Discovery and Matchmaking
In [11] algorithms were proposed to semantically classify and rank matches between
a request and available resources based on their logical descriptions. In a nutshell,
rankPotential algorithm allows to rank resources according to the degree of potential
satisfaction of a user request w.r.t. a resource when their descriptions are logically com-
patible, while rankPartial allows to obtain a ranking also when the resource and the
request are logically disjoint. Without delving into details we just mention that the orig-
inal framework has been adapted to our mobile ad-hoc scenarios based on semantically-
enhanced SDP. In our RFID setting a user request is built from the initial interest in a
specific resource; the system can suggest similar goods but also goods to be used in
combination with the picked up one. To this aim, a two-step discovery is performed,
exploiting two related ontologies. In the first step rankPotential algorithm cited above
is exploited to retrieve correspondences with the request and to identify similar re-
sources. Not compatible ones are ranked in the second step by means of rankPartial .
The hotspot will thus provide two different lists of records, respectively for resources
in a potential correspondence with the request and in a partial one. In advanced mobile
environments, the match between a request and a provided resource involves not only
the description of the resource itself, but also data-oriented contextual properties. An
overall utility function has to combine these values with matchmaking results, in or-
der to give a global match measure. In the proposed framework the utility function is
based on price (in US dollars), estimated delivery time (in days) and specific product
categories, as shown in Table 4. The utility function has a two-fold expression, for po-
tential and partial matches (similar resources and to be combined ones, respectively). In
both formulas the leading term is represented by the semantic match. A higher utility
function value corresponds to a better match.
4.2 Leveraging ONS for Ontology Support
The Object Naming Service (ONS) [16] mechanism is a supplementary system to grant
the so-called ontology support. Note that the whole proposed system is basically struc-
tured as a MANET. Hence, in case the reader does not manage the ontology w.r.t. the
tag annotation is expressed, it needs to retrieve the related DIG file via the Internet.
and WordCount respectively are the starting address and the number of memory words to be
read; if WordCount is 0, the tag will send all the memory words up to the end of the selected
bank.
30

Hence we use the ONS service planning to register within the EPCglobal Network Pro-
tocol Parameter Registry the new dig service suffix. It will indicate a service able to
retrieve ontologies with a specified OUUID value. Of course the same could be done
for OWL. In case of EPC derived from the GS1 standard [17], we reasonably assume
that the pair of fields used for ONS requests and referred to the manufacturer and to the
merchandise class of the good, will correspond to a specific ontology. In fact that pair
identifies exactly the product category. Two goods with the same values for that field
parameters will be surely homogeneous or even equal.
5 Case Study
An agent-based middleware integrating RFID and Bluetooth environments at the appli-
cation layer has been developed upon IBM WebSphere RFID Tracking Kit [18] to verify
the approach in a mobile marketplace setting. As mentioned above, a virtual consumer
electronics store was selected as case study. Annotations of products in the marketplace
is referred to a consumer electronics ontology, marked with a specific identifier we in-
dicate OU U ID
E
. The store hotspot performs semantic matchmaking of resource anno-
tations, exploiting a reasoner that executes rankPotential and rankPartial algorithms. In
the proposed approach we adopted MAMAS-tng [11]. Let us suppose Claire is looking
for a new laptop computer. She notices a quite cheap notebook model, bundled with an
office productivity suite. She puts it into the smart shopping cart. Sensor detects the cus-
tomer took a product. The RFID reader is triggered and reads data stored within the tag
attached to the laptop package, as described in Section 4. Then it is deactivated again.
Tag data consists of product EPC, ontology identifier OU U ID
E
, annotated semantic
description (stored as a DIG expression in a compressed encoding) and contextual pa-
rameters. Let us suppose that tagged description corresponds to a notebook with Intel
Centrino Core Duo CPU, 1 GB RAM, 80 GB hard disk drive, DVD writer and wireless
LAN connectivity; it has Windows XP Home Edition OS and an office software suite.
The equivalent expression in DL formalism is:
– notebook ⊓ ∀ has
CP U.Intel centrino core duo ⊓
∀ has
HDD.hard disk 80 GB ⊓ ∀ has disc recorder.DV D rec 16X 6X ⊓
∀ has
ram.ram 1 GB ⊓ ∀ has cards.wireless 802 11 card ⊓
∀ has
OS.W indows XP Home edition ⊓ ∀ has sof tware.suite of fice
w.r.t. OUU ID
E
reference ontology. Price is $550, delivery time is 0 days and product
category is 3.
The tablet touchscreen shows the received product details for building further se-
mantic based requests. Let us suppose Claire likes her choice. Now she would like to
find some basic accessories. She confirms the system-recommended request, which is
submitted via the semantic-enhanced Bluetooth SDP from the smart shopping cart to
the hotspot. Let us suppose the following products are available in the consumer elec-
tronics store knowledge base:
– S1: notebook with AMD Athlon XP-M CPU, 1 GB RAM, 80 GB hard disk drive, DVD
writer and wireless LAN connectivity. It is bundled with Windows XP Professional and an-
tivirus software. Price is $599; delivery time is 0 days; product category is 2:
31

Table 5. Matchmaking results.
Supply
Compatibility (Y/N)
rankPotential
score
rankPartial score
f(·)
S1: notebook with antivirus Y 6 - 0.001
S2: notebook with office suite Y 3 - 0.236
S3: desktop computer N - 79 0.166
S4: notebook bag N - 26 0.502
S5: UMTS phone N - 23 0.443
notebook ⊓ ∀ has CP U.AMD Athlon XP M ⊓ ∀ has HDD.hard disk 80 GB ⊓
∀ has
disc recorder.DV D rec 16X 6X ⊓ ∀ has ram.ram 1 GB ⊓
∀ has
cards.wireless 802 11 card ⊓ ∀ has OS.W indows XP P rof essional ⊓
∀ has
software.antivirus
– S2: notebook with Intel Centrino Core Duo CPU, 1 GB RAM, 80 GB hard disk drive, DVD
writer and wireless LAN connectivity. It is bundled with Linux and an office suite. Price is
$529; estimated delivery time is 1 day; product category is 2:
notebook ⊓ ∀ has CP U.Intel centrino core duo ⊓
∀ has
HDD.hard disk 80 GB ⊓ ∀ has disc recorder.DV D rec 16X 6X ⊓
∀ has
ram.ram 1 GB ⊓ ∀ has cards.wireless 802 11 card ⊓ ∀ has OS.Linux ⊓
∀ has
software.suite of f ice
– S3: a desktop computer with Intel Pentium 4 CPU, 1 GB RAM, 250 GB hard disk drive,
DVD writer, wireless LAN connectivity and a LCD display. It is bundled with Windows XP
Home Edition and an office suite. Price is $499; delivery time is 0 days; product category is
2:
desktop
computer ⊓ ∀ has CP U.Intel P entium 4 ⊓
∀ has
HDD.hard disk 250 GB ⊓ ∀ has display.LCD display ⊓
∀ has
disc recorder.DV D rec 16X 6X ⊓ ∀ has ram.ram 1 GB ⊓
∀ has
cards.wireless 802 11 card ⊓ ∀ has OS.W indows XP Home edition ⊓
∀ has
software.suite of f ice
– S4: a blue notebook bag. Price is $19; delivery time is 0 days; product category is 2:
notebook
bag ⊓ ∀ has color.blue
– S5: a silver-colored UMTS mobile phone with dual display and miniSD memory card sup-
port. Price is $169; delivery time is 0 days; product category is 1:
mobile
phone ⊓ ∀ has connectivity.U M T S ⊓ ≥ 2 has display ⊓ ≤
2 has
display ⊓ ∀ has display.LCD display ⊓ ∀ has memory card.mini sd
The hotspot performs the discovery and matchmaking processes as described in
Section 4.1 and returns results via Bluetooth SDP. Matchmaking results for this example
are presented in Table 5. The second column shows whether each retrieved resource is
compatible or not with request R. If yes, the rankPotential computed result is shown,
otherwise the rankPartial computed result is presented. In the last column results of the
overall utility function are reported.
Note that S2 is ranked as the best supply for similarity match, despite a longer
delivery time than S1. This is due to a better rankPotential outcome. Among candidate
resources for combination, category affinity favors S4 over S5, while S3 has a clearly
poorer match. For each retrieved resource a picture is provided along with matchmaking
score, price and description, as displayed in Figure 2. After finalizing her purchase,
Claire leaves the shopping cart in the store cart rack. This event is detected and the
tablet touchscreen returns to a quiescent state, waiting for another customer.
32

Fig.2. Retrieved resources are shown to the user.
6 Conclusion
In this paper we proposed a unified framework integrating RFID technologies with
enhanced Bluetooth Service Discovery Protocol supporting formal semantics. Objects
tagged with RFID transponders carry a semantically annotated description so permitting
to implement an advanced object discovery. Some slight modifications to the EPCglobal
standards have allowed the support to ontology-based data as well as to non standard
inference services, while keeping backward compatibility. The system has been imple-
mented within a message-oriented commercial middleware in order to test the feasibil-
ity and the usability of the proposed solution.
Acknowledgements
We wish to acknowledge support of Apulia project PE
074 “IC Technologies for track-
ing of agricultural and food products equipped with RFID tags”.
References
1. Weinstein, R.: Rfid: A technical overview and its application to the enterprise. IT Profes-
sional 07 (2005) 27–33
2. Ruta, M., Di Noia, T., Di Sciascio, E., Donini, F.: Semantic-Enhanced Bluetooth Discovery
Protocol for M-Commerce Applications. International Journal of Web and Grid Services 2
(2006) 424–452
3. Watson, R., Pitt, L., Berthon, P., Zinkhan, G.: U-Commerce: Expanding the Universe of
Marketing. Journal of the Academy of Marketing Science 30 (2002) 333–347
4. R
¨
omer, K., Schoch, T., Mattern, F., D
¨
ubendorfer, T.: Smart Identification Frameworks for
Ubiquitous Computing Applications. Wireless Networks 10 (2004) 689–700
33

5. De, P., Basu, K., Das, S.: An ubiquitous architectural framework and protocol for object
tracking using RFID tags. In: The First Annual International Conference on Mobile and
Ubiquitous Systems:Networking and Services (MOBIQUITOUS 2004). (2004) 174–182
6. Hum, A.: Fabric area network a new wireless communications infrastructure to enableu-
biquitous networking and sensing on intelligent clothing. Computer Networks 35 (2001)
391–399
7. Schmidt, A., Gellersen, H., Merz, C.: Enabling implicit human computer interaction: a wear-
able RFID-tagreader. In: The Fourth International Symposium on Wearable Computers.
(2000) 193–194
8. Siegemund, F., Florkemeier, C.: Interaction in Pervasive Computing Settings using
Bluetooth-Enabled ActiveTags and Passive RFID Technology together with Mobile Phones.
In: Proceedings of the First IEEE International Conference on Pervasive Computingand
Communications (PerCom 2003). (2003) 378–387
9. Kawakita, Y., Wakayama, S., Hada, H., Nakamura, O., Murai, J.: Rendezvous enhancement
for conference support system based on RFID. In: International Symposium on Applications
and the Internet Workshops (SAINT2004). (2004) 280–286
10. Traub, K., Allgair, G., Barthel, H., Bustein, L., Garrett, J., Hogan, B., Rodrigues, B., Sarma,
S., Schmidt, J., Schramek, C., Stewart, R., Suen, K.: EPCglobal Architecture Framework.
Technical report, EPCglobal (2005)
11. Di Noia, T., Di Sciascio, E., Donini, F., Mongiello, M.: A System for Principled Matchmak-
ing in an Electronic Marketplace. International Journal of Electronic Commerce 8 (2004)
9–37
12. W3C Recommendation: OWL Web Ontology Language. www.w3.org/TR/owl-features/
(2004)
13. Bechhofer, S., M
¨
oller, R., Crowther, P.: The DIG Description Logic Interface. In: Proceed-
ings of the 16th International Workshop on Description Logics (DL’03). Volume 81 of CEUR
Workshop Proceedings. (2003)
14. Baader, F., Calvanese, D., Mc Guinness, D., Nardi, D., Patel-Schneider, P.: The Description
Logic Handbook. Cambridge University Press (2002)
15. Epcglobal Inc.: EPC Radio-Frequency Identity Protocols Class-1 Generation-2 UHF RFID
Protocolfor Communications at 860 MHz–960 MHz. Technical report, EPCglobal (2005)
16. EPCglobal Ratified Specification: Object Naming Service (ONS - ver. 1.0).
http://www.epcglobalinc.org (October, 4, 2005)
17. EPCglobal: (http://www.epcglobalinc.org)
18. Chamberlain, J., Blanchard, C., Burlingame, S., Chandramohan, S., Forestier, E., Griffith, G.,
Mazzara, M., Musti, S., S.-Y.Son, Stump, G., Weiss, C.: IBM WebSphere RFID Handbook:
A Solution Guide. IBM International Technical Support Organization (2006)
34