Using Active and Passive RFID Technology to ... - Semantic Scholar

578

IEEE Transactions on Consumer Electronics, Vol. 54, No. 2, MAY 2008

Using Active and Passive RFID Technology to Support Indoor Location-Aware Systems Tesoriero, R., Gallud, J. A., Lozano, M., and Penichet, V. M. R.

Abstract — Emerging mobile technologies are being applied successfully in cultural spaces to help users to get the correct information no matter where he is, in such a way that their use result better than traditional guides. Different technologies (RFID, Wi-Fi, etc.) are applied to allow mobile devices, such as PDAs or smart phones, interact with their environment in order to provide users with updated information regarding his physical location. This paper describes a system environment based on both active and passive RFID technology which supports the automatic positioning of mobile devices in closed spaces. This proposal results especially useful to provide location-aware information avoiding the need for users to manually select the desired information they want to view. This proposal has been successfully proved for the location and positioning of PDAs1. Index Terms — active and passive RFID, location-aware systems, mobile devices, indoor automatic location.

I. INTRODUCTION Positioning systems are common in software applications that manage geographic information. Indeed, using computing devices while on the move is currently very common. PDAs, mobile phones, car navigation systems, and mobile entertainment devices are some examples. The basic idea of location and context-awareness is to provide services that fit the user's current situation. An interesting challenge is how to locate a mobile device inside a building as in the case of administrative buildings, hospitals or art museums. The idea is creating systems that continuously adapt to the user's needs and surroundings. This paper describes an indoor location-aware approach to support the inclusion of automatic positioning of mobile devices based on a well known technology such as RFID (Radio-Frequency Identification). Previous proposals show the interest of the application of context-awareness in museums and other public spaces [1, 2,

4]. Our proposal employs both active and passive RFID jointly with other known technologies (mobile devices, infrared and Wi-Fi). The proposal is implemented in a real scenario and the paper shows also the evaluation of the prototype. The rest of the paper is organized as follows: Section 2 describes the state of the art regarding location-aware information. In Section 3, an overview of the RFID technology is presented. Section 4 explains the positioning system based on RFID proposed in this paper. Section 5 describes some practical issues and the evaluation of the system. Finally, some conclusions and final remarks are presented in Section 6. II. SUPPORTING LOCATION-AWARE INFORMATION A. Requirements of location-aware information systems Our interest is focused on the requirements, techniques and technology needed to support indoor automatic positioning, a subset inside the wide field of location-aware systems. Indoor location-aware systems have specific requirements, specially the precision of the location which is over 30cm and 1mtr. In this case we should talk about positioning systems, that is, the functionality dedicated to maintain the user’s position within a building. Positioning systems can provide absolute or relative coordinates of the user. This information can be managed either in a centralized or distributed system [7, 11]. Precision is the metric used to compare different indoor positioning systems. These systems should be scalable and adaptable, in the sense of the ability to accept external changes. In the case of mobile devices or mobile positioning systems, the power source is an important element, as we will see later in this paper. The environment has also an influence on the precision and scalability of the positioning systems. Finally, the higher precision, scalability and adaptability a positioning system requires, the higher it will cost.

1 This work has been partially supported by JCCM funds under grant “PCC05-005-1” and the Spanish Science and Education Ministry funds under B. Indoor positioning techniques grant “TIN2004-08000-C03-01”. R. Tesoriero belongs to the Albacete Research Institute of Informatics, Among the most known indoor positioning techniques, the University of Castilla-La Mancha, 02071 Albacete, SPAIN (e-mail: first group is based on the electric signal: trilateration, [email protected]). proximity, multilateration, signal power J. A. Gallud belongs to the Computer Systems Department, University of A second group is based on dead-reckoning sensors: these Castilla-La Mancha, 02071 Albacete, SPAIN (e-mail: [email protected]). M. Lozano belongs to the Computer Systems Department, University of systems calculate the walking distance and the direction of the Castilla-La Mancha, 02071 Albacete, SPAIN (e-mail: [email protected]). user. These techniques can be used in combination with V. M. R. Penichet belongs to the Computer Systems Department, inertial systems to implement navigational inertial systems. University of Castilla-La Mancha, 02071 Albacete, SPAIN (e-mail: [email protected]). Contributed Paper Manuscript received November 22, 2007 0098 3063/08/$20.00 © 2008 IEEE

R. Tesoriero et al.: Using Active and Passive RFID Technology to Support Indoor Location-Aware Systems

A third group of positioning techniques use images as the features extraction, gray level intensity, edge detection and forms or object reckoning. In the next section we describe the technique we use in our approach. C. Indoor positioning technologies Previous sections describe the specific requirements and the techniques to develop an indoor positioning system. This section enumerates the main technologies that can be used to implement this kind of location-aware system oriented to provide a position in a building. The most important technologies are based on radio frequency, as in the case of Wi-Fi (IEEE 802.11x) or RFID (Radio Frequency Identification). There are some solutions based on Wi-Fi [8] which employ different techniques to automatically locate the user in an indoor environment. Nevertheless, indoor Wi-Fi performance is not good enough and the implantation cost is too high. Other radio frequency based technology to be employed could be Bluetooth (IEEE 802.15). The special features of Bluetooth technology do not suit well with the real-time needs of positioning systems. Infrared and ultrasounds have been also used to implement indoor positioning systems. Together with dead-reckoning techniques, accelerometers, digital compasses, barometers and so on, can be used to build navigational inertial systems. RFID is the chosen technology in our proposal due to its special characteristics and it will be described in the next section. D. Indoor positioning systems In this section, the main indoor positioning systems are presented. These systems offer a solution integrating some of the techniques and technologies previously described. The first group makes use of Wi-Fi technology. One of the first systems locates users within a building, and uses infrared technology [8]. Other employs the signal strength to locate a user with 5 meters of precision [8]. A third solution uses access points and 802.11b tags together with a high cost software client. Finally, there is a system that uses special devices to connect with the 802.11 clients. We have decided not to use indoor positioning based on Wi-Fi technology because of the problems it has to get a good precision. There are a few real solutions based on RFID technology. One of them is used far different from our proposal as they employ mobile RFID tags whereas we propose using mobile RFID readers. III. THE RFID TECHNOLOGY RFID technology is replacing barcodes in those applications that need to identify an item, like product tracking, inventory systems, libraries, and so on. The RFID technology employs an electronic element called tag or transponder to save an identification number which is retrieved by using an RFID

579

reader. The reader device can be managed from a mobile or a desktop computer. RFID is successfully being applied to a variety of fields and new applications of this technology are constantly appearing [3, 5 and 14]. The RFID tags can be passives or actives depending on the way the signal is induced. Passive tags do not incorporate power supply because the induced electrical current by the reader is enough to transmit the response. The lack of battery in the tags has an important influence on their size that could be extremely reduced.

Fig. 1. Examples of active and passive RFID tags and readers for mobile devices

Active tags use their own internal power source to broadcast the signal. The main difference between active and passive RFID technology regards with the reading distance. Using active RFID is possible to detect a RFID tag at a hundred of meters. Figure 1 shows both active and passive RFID tags and a PDA including a RFID reader. Some applications need to know the real position of a mobile user to provide him with some contextual information. The use of RFID technology in these scenarios is relatively new, especially in the case of providing indoor location awareness [10]. IV. A RFID APPROACH TO SUPPORT INDOOR LOCATIONAWARENESS

In this section we describe our proposal based on RFID technology to support an indoor location-aware system. The positioning system has been developed in a real scenario, an example of an indoor building with special need of positioning as it is an art museum [4, 9, 13]. Figure 2 shows the server infrastructure of the deployed system. The PDA Wi-Fi network is employed by users running a mobile application on a PDA.

580


installed in the museum and which contains the pieces catalogue and, eventually, a media repository). The external database refers to our specific database to support the positioning subsystem, the information to be displayed to visitors, internationalization issues and so on. When a user with a PDA visits the museum, he or she can download and run the client program at the entrance hall. In order to retrieve the information related to exhibited pieces, the user may choose between auto-navigation mode or manual navigation mode. Manual navigation mode allows users to navigate across the building using cursor keys or the PDA stencil. On the other hand, the PDA is able to detect automatically the position of the user in the museum and retrieve the correct information according to the PDA location in the museum reducing considerably the user clicks on the PDA. Fig. 2. Real scenario infrastructure. The figure shows the server side and the Wi-Fi networks

The visitors of the museum use a mobile device to interact with the environment by means of RFID technology. One of the first tasks to be performed is the definition of a conceptual model that incorporates location information into the system. A cultural space entity is conceptually composed by, at least, two types of information, the Catalogue information and the Environmental information. The important part here is the Environmental information. An art object or piece being exposed is wrapped by extra information that depends on the exposition environment; for example the physical space in which the piece is placed. So, the central item of our model is the Space. A piece must be exposed into a Space and it has a Space associated to it too. Pieces are usually exposed within a container, for instance a show. A show may represent a showcase or a frame, in case of a painting, or anything that is able to contain a piece. A Space has a graphical representation, associated resources, and a set of identifiers. An interesting thing to point out here is the fact that the Identifier entity isolates the system from the specific technology (RFID, Wi-Fi, etc) used to locate any object in the real world. Thus, it is possible to use different positioning technologies at the same time to improve the system accuracy. For instance, you can use Active RFID tags to identify a showcase and code bars or Passive RFID tags to identify a piece into the showcase. The desired functionality for the system includes the use of context aware information to provide a richer user experience. Additionally, we integrate our solution to a software application currently running in the museum (the catalogue application). The positioning subsystem is responsible for giving the PDA an identification to locate the device according to a relative or absolute position. The automatic database synchronization subsystem is responsible for maintaining the coherence between both the internal and external databases. As internal database we mean the database that may be already

Fig. 3. Active (location) and passive (gesture) RFID schema. The RFID reader is installed on a PDA

Figure 3 shows the two different mechanisms used to implement the context-awareness system. The variety of positioning systems forced us to define the system independently from the hardware employed to locate users. Thus, the client program running on the PDA can receive information from the environment in many different ways: infrared sensors, active or passive RFID tags, Bluetooth, Wi-Fi devices or any other system available now or to appear in the future. So, the client sends this information to the server and receives back the requested information. The main advantage of this system is the possibility to interact with the environment in three levels of attention. The lower level of attention allows us to perceive the periphery without requiring hardly attention from the user. So the user can perform other activities without being interrupted;


but, he or she can be aware of other interesting events in case they occur. For instance, if we are visiting a gallery and we are talking while walking across a corridor. In such case, we may come across an important masterpiece without noticing and we would miss it. However, if we were notified on time, we would not have missed it. To solve this problem, we use active RFID tags to perform this task. When the PDA is close to an interesting point, the user is notified about it and the related information is retrieved and displayed on the PDA. The medium level requires more attention from the user because he or she has to perform an explicit action to retrieve the information regarding a piece or space. For instance, if we are watching a showcase and we want to get some information about a specific piece, then we can put the device near the piece label and retrieve extra information about it. This technique uses passive RFID tags that are detected by the reader at 5-8 cm from the tag. The higher level requires even more attention than its predecessors, and the user should point to an object or label in order to retrieve information about it. In this last case, the interaction is not as natural as in the other alternatives, but it provides a good method when information is not easily reachable because of the distance. For instance, suppose that we want to retrieve information about an artifact that hangs from the roof. The user may point to the object and retrieve information. To implement this alternative, IR technology is used and the user has to point the mobile device to the object to get the information. The system works in this way. Mobile devices interact with their environment in two different levels of granularity: coarse grained and fine grained. In coarse grained mode, the system is able to perceive the location of the user relatively to a position. To perform this task active RFID tags are the main actors. The environment provides mobile users with the information of the periphery and mobile software detects art objects (for instance, a statue or a masterpiece) and asks users to download additional information about it. Fine grained mode is used to identify objects. In this case we may use passive RFID or infrared according to the situation. An example of this situation is the selection of a piece among a set of them placed in a showcase. However, if passive and active RFID are used simultaneously, a hardware problem appears. Most of the current commercially available PDAs have only one Compact Flash slot, if any; and RFID readers are only available in this format for PDA. Therefore, in this case we have to use a Compact Flash hub in order to get active and passive RFID readers working simultaneously. This solution also improves the portability of the application. In order to achieve it, we established an architecture to support portability among mobile devices as general as possible. Interface portability is an important issue to take into account when designing software for portable devices. The

581

main reason is the fast development that technology has in this area. Mobile devices evolve quickly and new PDAs, Smartphones or other mobile devices are turning into the desktop offices. Mobile devices exchange information with the server using XML through an HTTP connection. The server receives requests and fetches the information from the database replying back to the client the required information. Then, this information is displayed on the client screen. V. PRACTICAL ISSUES The case of study where this technology has been applied is an art museum, where the art objects are cutlery (knifes, pocket knifes, scissors and so on). There is a mobile application running on 24 PDAs that are offered to visitors at the entrance hall of the museum and a prototype developed in our research group. This prototype includes the automatic positioning system based on RFID technology. A. Preparing the museum prototype in the research lab In order to evaluate the precision, exactitude, scalability and energy features of our system, we designed a set of simulated showcases and art pieces. Figure 4 shows an example of a specific showcase simulating a real one.

Fig. 4. Simulated showcase and art pieces. The title of the show explains that the user is viewing Old Spanish Pocket Knifes

In Figure 4 there is a special icon next to each piece representing the passive RFID tag. The passive RFID tag is located below its representation. The showcase of Figure 4 has 16 passive RFID tags and an active RFID tag. Each room in the simulated museum has an active RFID tag. When the user enters into a room, a map representation of the room is automatically displayed on the PDA. The information corresponding to the different art pieces may be seen in two different views: catalog information and environmental information. The catalog information view is related to technical issues about pieces in the museum (i.e. maintenance dates, revisions,

582

value, and so on) and is written using technical language. On the other hand, the environmental information view is intended for visitors (i.e. related audios and videos, contextual information, etc) and colloquial language is used in this case in order to improve users’ experience and understandability. Although environmental information is provided to visitors in general; technical information may be extremely useful for certain users that may require more information about some concrete pieces in the museum. Figure 5 shows an example of the user interface.


Format for Usability Reports) standard defined by the ISO/IEC DTR 9126-4 [6]. The questionnaire is the following: Do you like the application design? (1) Do you like the application appearance? (2) Do you feel graphics has good quality? (3) Is this application easy to learn? (4) Did you get a good feeling about using the application? (5) Was it difficult to choose the language? (6) Is it easy to choose a room? (7) Is it easy to get a showcase? (8) Is it easy to find a piece? (9) What do you think about usability? (10).

Fig. 6. Satisfaction results by question. Fig. 5. The RFID reader automatically detects a showcase and art objects are displayed on the PDA.

B. Evaluating the system A first version of the system has been running in a real museum in Albacete, Spain since September 2004 and during these 3 years, the museum has received over 15.000 visitors and over 3.000 visitors have used the system. An upgraded version of the system will be deployed in autumn 2007. The users’ satisfaction tests carried out with real users proved the validity of the first approach of our system. A summary of these results is described next. This section does not try to show all the results or their analysis but an overall view of the users’ opinion. The evaluation was performed on an HCI Lab simulating the museum showcases as it is explained in the previous section [12]. Although users were familiar with computing systems (100%), they are not so with mobile devices such as PDAs. The amount of people that filled the questionnaire was 14. The experiment embraces three aspects of the application that will be evaluated to get the users’ experience satisfaction level: • Aspect 1: Graphic design and visual impact (questions 1, 2 and 3). • Aspect 2: General usability (questions 4, 5 and 10). • Aspect 3: Concrete functionality (i.e. finding a showcase). To retrieve user feelings about the application and the use of RFID technology to provide the services previously described, we used a questionnaire following the CIF (Common Industry

The results were analyzed according to the percentage of satisfaction in two ways: by question and by aspect [12]. Figure 6 shows the satisfaction by question. The average of satisfaction index is 75.95% and it is significant if we take into account that 100% of people are familiar with PCs. C. Other technical issues We have employed PDA with special batteries, passive reader and active reader with especially suitable for PDA. The application has been developed using C# and we have had to develop our own libraries to work with the RFID readers. In the case of active RFID tags, the technique used to locate the user in the space depends of the signal strength. These tags use a battery and the signal level is directly correlated with the battery level. This forced us to design our power source based on 9v batteries. This power source allows us to determine precisely the strength of the signal according to the features of the building. VI. CONCLUSIONS AND FUTURE WORK In this paper we present an approach to implement an indoor location-aware system. This approach uses active and passive RFID technology and mobile devices (PDAs) implemented in a real scenario, and shows the usefulness of this technology to support automatic positioning. The paper introduces a location system that is able to both locate users and identify objects at the same time combining different technologies (IR, RFID, etc.) to get three levels of


attention which produces interaction at different space granularities. The system provides users with new ways of interaction that replaces navigation based on clicks performed by the user to determine his or her location in the system. The location aware system proposed in this paper has been implemented using a simulated real scenario and the prototype has been evaluated with real users. ACKNOWLEDGMENT Special thanks to the Cutlery Museum of Albacete (Albacete, Spain) and all the people of the Computer Science Research Institute for collaborating in the evaluation of the system. REFERENCES [1]

Barber, C., Bristow, H., Cheng, S., Hedley, A., Kuriyama, Y., Lien, M., Pollard, J., Sorrell, P. (2001). Augmenting Museums and Art Galleries. Human-Computer Interaction INTERACT '01, The International Federation for Information Processing, Tokyo, Japan, 439-447. [2] Ciavarella, C., Paternò, F. (2004). The design of a handheld, locationaware guide for indoor environments. Personal and Ubiquitous Computing, Vol.8 N.2, Springer Verlag, 82-91. [3] Floerkemeier, C., Lampe, M. “Middleware: RFID middleware design: addressing application requirements and RFID constraints “. October 2005 Proceedings of the 2005 joint conference on Smart objects and ambient intelligence: innovative context-aware services: usages and technologies sOc-EUSAI '05 [4] Grinter, R.E.; Aoki, P.M.; Hurst, A.; Szymanski, M.H.; Thornton J.D. and Woodruff, A. (2002). Revisiting the Visit: Understanding How Technology Can Shape the Museum Visit. In Proc. ACM Conf. on Computer Supported Cooperative Work, New Orleans, LA. [5] Hori, T., Matsumoto, M. “The challenges of wireless and mobile technologies: the RFID encourages the mobile phone development”. October 2004 Proceedings of the 3rd international conference on Mobile and ubiquitous multimedia MUM '04 [6] ISO/IEC TR 9126-4:2004. Software engineering.Product quality.Part 4: Quality in use metrics. JTC 1/SC 7; ISO Standards. ICS: 35.080. Stage date: 2004-03-17. [7] LaMarca, A.; Chawathe, Y.; Consolvo, S.; Hightower, J.; Smith, I.; Scott, J.; Sohn, T.; Howard, J.; Hughes, J.; Potter, F.; Tabert, J.; Powledge, P.; Borriello, G. and Schilit, B. (2005). Place Lab: Device Positioning Using Radio Beacons in the Wild. Intel Research Technical Report: IRSTR-04-016. In proceedings of Pervasive 2005, Munich, Germany [8] Lemieux, N. “Reading course on Indoor Location Systems”. http://www.csd.uwo.ca/~nplemieu/readingCourse/IndoorLocationReadi ngCourse.html [9] Long, S., et al. (1996). Rapid Prototyping of Mobile Context-aware Applications: The Cyberguide Case Study. 2nd ACM International Conference on Mobile Computing and Networking (MobiCom'96) 1996 November 10-12, 1996. [10] Ni, L. M.; Liu, Y.; Lau, Y. C. and Patil, A. P. (2003): LANDMARC: Indoor location sensing using active RFID. Proceedings of the First IEEE International Conference on Pervasive Computing and Communications, 2003. (PerCom 2003). Pág 407- 415. ISBN: 0-76951893-1. [11] Ravi, N.; Shankar, P.; Frankel, A.; Elgammal, A. and Iftode, L. (2006): Indoor localization using camera phones, Proceedings of the 7th IEEE Workshop on Mobile Computing Systems and Applications, April 2006. [12] Tesoriero, R.; Lozano, M. D.; Gallud, J. A. and Penichet, V. M. R. (2007c): Evaluating the Users’ Experience of a PDA-Based Software Applied in Art Museums. In Proceedings WebIST 2007. March 3-6, 2007. Barcelona, Spain

583

[13] Tesoriero, Ricardo; Montero, Francisco; Lozano, María Dolores and Gallud, José Antonio (2007b): HCI Design Patterns for PDA Running Space Structured Applications. Publication pending HCI International 2007. . Beijing, China. 22-27-Julio, 2007. [14] Yoder, R. C. “Using RFID in the classroom to teach information systems principles”. June 2006, Journal of Computing Sciences in Colleges, Volume 21 Issue 6 Ricardo Tesoriero was born in Argentina on July 1st, 1977, he got his degree in Computer Science at National University of La Plata n 2005 and he is actually performing is PhD studies at University of Castilla-La Mancha in Spain. From 1999-2005 he was working at LIFIA (Research and development in Advanced IT) and from 2005 until today he is working at LoUISE (Laboratory of User Interaction and Software Engineering) at Albacete Research Institute of Informatics. He was awarded with Joaquín V. Gonzalez distinction in 2003. J. A. Gallud Jose A. Gallud was born in Valencia (Spain) on September 9st, 1966; he got his degree in Computer Science at Universidad Politécnica de Valencia in 1991 and the PhD degree at University of Murcia in 2001. He is full professor of Computer Sciences at University of Castilla-La Mancha in Spain since 1993. His initial research focused on parallel computing and the development of distributed applications. Since 2002 he belongs to the Laboratory of User Interaction and Software Engineering (LoUISE) research group at the University of Castilla-La Mancha. He is responsible of different research projects about the development of mobile applications and context-awareness systems. M. Lozano. MSc and PhD in Computer Science by the Technical University of Valencia, Spain. Currently she is Associate Professor at the Department of Computer Systems in the University of Castilla-La Mancha. She belongs to the Laboratory of User Interaction and Software Engineering (LoUISE) research group of the Albacete Research Institute of Informatics. Her teaching and research areas concerns Software Engineering and Human Computer Interaction. She is author of numerous papers and member of several Program Committees of different national and international conferences. Her research interests are, among others, Model-based User Interface Development, Software Engineering, Usability, CSCW, etc. V. M. R. Penichet was funded by Junta de Comunidades de Castilla-La Mancha to do his Ph. D. studies (20042007). From July 2007, he is Assistant professor at the UCLM teaching at the Superior Polytechnic School of Albacete (EPSA). He belongs to the Laboratory of User Interaction and Software Engineering (LoUISE) at the same University. His current research is about ComputerSupported Cooperative Work (CSCW) and Human-Computer Interaction (HCI), and its methodological aspects according to the Software Engineering. As a result of his research there are some publications in several national and international conferences. From June 2005 to January 2006 worked at the Institute of Information Science and Technologies (ISTI) at the Italian National Research Council (CNR) with Dr. Fabio Paternò. The stay was funded by the European ADVISES TMR.