Indoor Localization System for First Responders in Emergency Scenario

Transcription

1 Indoor Localization System for First Responders in Emergency Scenario Romeo Giuliano +, Franco Mazzenga +, Marco Petracca, Marco Vari + + University of Rome Tor Vergata, Department of Enterprise Engineering, Rome, Italy Radiolabs, Consorzio Università Industria, Rome, Italy Contact author: marco.petracca@radiolabs.it Abstract Recently, wireless indoor location systems have been successfully used in many applications, such as asset tracking, indoor navigation, and inventory management. Such systems become essential in Public Safety and Disaster Relief (PSDR) scenarios, where a robust and reliable first responder positioning system can significantly increase safety in emergency operations. In this paper, we characterize the performance of a RFID-based indoor localization system. We present various tests aimed at assessing the capability of the ILS to identify first responders in different conditions. Results show that, by means of a correct dimensioning of the ILS, high efficiency can be achieved in localization. Keywords RFID; indoor localization; PSDR; measurements I. INTRODUCTION In recent years, international research increased the focus on systems for indoor localization, namely Indoor Location Systems (ILSs) [1]. Global Navigation Satellite Systems (GNSSs) such as GLONASS (Global Navigation Satellite System) and GPS (Global Positioning System) [2] are widely used in outdoor environment thanks to the large coverage. Nevertheless, they can not be used in indoor environments due to low received power from the satellites. Conversely, an ILS enables mobile devices to determine their indoor position, that could be exploit for location-based services, such as navigation, tourism, tracking, monitoring, etc. However, it should be considered that indoor environments are generally more complex than outdoors. There are different obstacles, such as walls, equipment, people, etc., which can affect the propagation of electromagnetic waves, leading to multi-path effects and high attenuation. In addition, noise and other interference sources such as wireless and wired networks can cause reduction in localization accuracy. Finally, the structure of the building, the mobility of people and the critical environmental conditions (e.g. high temperature and high pressure) can impact on the system used for localization [3]. Many applications related to indoor localization assume vital importance in Public Safety and Disaster Relief (PSDR) scenarios, where a robust and reliable first responder positioning can significantly increase safety in emergency operations. Indeed, one of key aspect during emergency situations is the knowledge of the location of rescue teams. Accurate localization will increase the level of safety and reduce the mission time, as rescuers can be better coordinated and guided /13/$31.00 c 2013 IEEE In addition, a reliable ILS would reduce the rescue team disorientation, that strongly contribute to the mortality rate of the rescuers, as reported by the U.S. National Institute for Occupational Safety and Health (NIOSH) [4]. In some cases, for example, the firefighters have only few seconds to get to safety, having to find the exit as fast as possible and often without being able to follow the same path used to enter the building, due to collapsed ceilings or floors. Even in non-time critical situations the rescue teams should be able to find the generic firefighter inside a building. Moreover, the availability of the building layout can be an useful information for the rescue teams. Another important issue for ILS is the operation & maintenance aspect. The positioning support system should be not only installed but also maintained, since reference elements in the systems (e.g. radio bearers, tags or ultrasound sensors) can be damaged, stolen or can simply expire their life time. Moreover, the information used for localization should be as much universal as possible, since the ILS can be installed in public structures, commercial buildings or private houses, reducing the possibility of intervention on ILS by the rescue teams or other teams dedicated to positioning system maintenance. In this paper, we propose the implementation of an ILS based on RFID (Radio Frequency Identification) technology. The system uses passive RFID tags deployed in the indoor environment. Each of them has its pre-stored fixed location data, e.g. Universal Transverse Mercator (UTM) coordinates and building floor height. The first responder is equipped of an RFID Reader able to detect the RFID tag message that indicate him/her you are here. This way the first responder has the information not only in which floor but also in which room he/she is, due to limited RFID range. In particular, we characterize the system by carrying out various tests aimed at assessing the capability of the RFID to support first responder operations in different environmental conditions. The obtained results can be useful for ILS dimensioning, ensuring both resilience and a meter-level accuracy. The possible application of algorithms for efficient management of localization service in the area [5]- [8] is not considered in this work. The paper is organized as follows. In Section II we provide the rationale of this work and an overview of the related literature. The considered scenario is described in Section III. In Section IV we detail the experimental setup. In Section V the test planning is defined. The test results are reported and analyzed in Section VI. Finally Conclusions are drawn.

2 II. RATIONALE AND RELATED WORKS Many wireless indoor positioning solutions exist adopting different techniques and systems [1]. The three main location estimation schemes are triangulation, scene analysis and proximity, whereas many wireless technologies have been used for ILS, such as Infrared (IR), Wireless Local Area Network (WLAN), Cellular-based [9] [10], Wireless Sensor Networks (WSN), Ultra-wideband (UWB) [11]- [14] and RFID [15] [16]. This latter presents several advantages over the other technologies, like easier deployment and maintenance, high security and robustness, cost effectiveness, no need for synchronization cables (as required, for example, by UWB for implementing TOA methods), higher flexibility than other preinstalled devices (e.g., Access Points in WiFi networks), availability in non-line-of-sight (NLOS) propagation environments. For RFID systems, the proximity localization scheme is privileged, especially with passive tags or sparse reader deployment. This technique is very basic and relatively easy to implement. Generally, it provides localization accuracy comparable with the size of the cells. It is worthwhile to outline that the choice of the technique and the technology significantly impacts on the reliability and the accuracy of the location information but also on the cost and the efficiency of the ILS. By taking into account the considered PSDR scenario and the related system requirements, in this work we focus on a RFID ILS using passive tags. In the literature, many papers deal with RFID-Based indoor positioning. However, to the best of our knowledge, the robustness and efficiency of RFID ILS with respect to the specific PSDR scenarios and requirements have not been extensively investigated. In [17], the location estimation error is experimentally evaluated but considering static positions and relatively limited distances between tags and reader. In [18] authors provide results from real measurements that show the accuracy of the proposed system, but new dual function semipassive components, referred to as sensatag, are used. The integration of RFID tags, placed in the building as navigation waypoints, with inertial navigation systems has been also examined in [19], [20] to limit the well known drift problem of dead-reckoning position estimation methods. However, this hybrid approach does not reflect our considered scenario, since it requires an inertial measuring unit (IMU). III. SCENARIO The reference scenario is depicted in Fig.1. The RFIDbased ILS is represented by both the set of deployed RFID tags and the RFID reader. RFID tags are fixed inside a building in strategic positions, while the RFID reader on the first responder is mobile. This is the flipside of the typical RFID applications, which envisage mobile tags and fixed readers. The ILS is designed to be lightweight, small, low-cost and power efficient. It should be also able to still provide a meterlevel accuracy during extended operations. The RFID-based ILS is designed to reduce the dependence on wireless links to external data sources by exploiting the capability of RFID tags to store critical building information for local retrieval. As an example, upon interrogation each Tag can reply with the UTM coordinates and the floor number/height. Other data can be stored into the RFID bank memory to provide first responders with further local information (e.g., building structure, room type, presence of dangerous materials). Anyway, Fig. 1. Scenario. the RFID reader is endowed with bluetooth connectivity to transmit received information on detected tags to an User Device, like smartphone or tablet (note that other connectivity strategies can be adopted [21], eventually considering energy consumption constraints [22]). This terminal is able to retransmit positioning information to a Command and Control Centre (CCC) by means of 2G/3G/4G wireless networks, e.g. Public Land Mobile Networks (PLMNs) and WiMAX [23] [24], or Professional Mobile Radio (PMR), like TETRA [25]. In such a way, the CCC can collect and process positioning data in order to track and guide rescue workers during missions within unknown sites.it is worthwhile to note that the User Device can exploit the available embedded chipset to calculate its outdoor position (through A-GPS) as well as to integrate the indoor localization procedure (through Accelerometer Sensors, Digital Compass, etc.). This approach is out of the scope of this work and can be investigated in further studies. IV. EXPERIMENTAL SETUP Extensive measurements have been carried out to characterize the behavior of the RFID-based ILS. In Fig.2 we show the generic considered test bed. The main parameters that describe the system are the distance d between the Reader and the Tag and the azimuth and elevation angles for both devices, referred to as θ and φ, respectively. Indeed, besides the maximum operative distance, it is necessary to characterize the antenna radiation patterns of both the devices to fully understand their behavior. According to the functioning of the Fig. 2. Generic test-bed.

3 RFID system, the Reader transmits an interrogation. If the Tag receives enough power through the interrogation message, it re-transmits its code stored in its internal memory. Finally, the Reader can receive the Tag code if sufficient power is detected by its antenna. In such a case, the communication between the Reader and the Tag is successfully performed. The experimental setup is depicted in Fig.3. We utilize UHF Fig. 3. Experimental setup. passive Omni-ID Ultra Long Range RFID Tags [26]. They operate in the MHz frequency band and in the temperature range -40 C +65 C. As for the memory banks, these Tags have 96 bits Electronic Product Code (EPC) and 512 bits for the User Memory. The RFID CAEN A528 OEM UHF multiregional compact Reader [27] has a maximum transmission power of 500 mw on the MHz frequency range, with temperature comprised between -20 C and +60 C. According to the reference scenario, this reader is endowed with Bluetooth connectivity. The software package utilized to collect data and evaluate the performance of the RFID-based ILS is the CAEN RFID EASY2READ EASY CONTROLLER [28], which allows to explore the main CAEN RFID readers capabilities. Thanks to a friendly user interface, the user can read and inventory tags. In particular, this software is able to collect data related to the following parameters: a = tags/s, i.e., the average number of tag detections per second; b = acquisitions/s, i.e., the average number of reader queries per second, included the unsuccessful acquisitions (i.e., message tags not found is displayed); η = a/b, i.e., the efficiency of the RFID system; RSSI, i.e., the average value of the proportional received signal strength indicator; count, i.e., the number of tag detections over the test time. The EasyController application is also available for Android devices, allowing smartphones and tablets to directly communicate with the RFID Reader and perform data acquisition, in accordance with the envisaged scenario. V. TEST DESCRIPTION In this section we describe the test planning adopted to evaluate the system performance. We identify the following test classification: I) Class I - Ideal conditions: the tests of Class I are carried out to characterize the behavior of the ILS under ideal operative conditions, i.e., LOS transmissions between RFID Tag and Reader, regular environmental parameters (e.g., temperature, humidity, etc.), RFID reader set on a RF-transparent pole, etc. This test class, which represents the reference case, is performed in the scenario reported in Fig.4 and includes the following tests: Test I.1: evaluation of the operative distance of the RFID system. The test is carried out by measuring the system efficiency η as a function of the distance between Tag and Reader, facing one other (θ = φ = 0). In such a way, the maximum operative distance could be defined as the distance between devices such that the system efficiency is not null (i.e., η > 0 ) or above a predefined threshold (i.e., η η thr ); Test I.2: characterization of the elevation radiation pattern of the Tag antenna. The test is carried out by measuring the system efficiency obtained for various elevation angles of the Tag antenna positioned at increasing distances from the Reader; Test I.3: characterization of the azimuth radiation pattern of the Tag antenna. The test is carried out by measuring the system efficiency obtained for various azimuth angles of the Tag antenna positioned at increasing distances from the Reader; Test I.4: characterization of the elevation radiation pattern of the reader antenna. The test is carried out by measuring the system efficiency obtained for various elevation angles of the Reader antenna positioned at increasing distances from the Tag; Test I.5: characterization of the azimuth radiation pattern of the reader antenna. The test is carried out by measuring the system efficiency obtained for various azimuth angles of the Reader antenna positioned at increasing distances from the Tag; Fig. 4. Setup configuration for Class I Tests. II) Class II - Typical configuration: this type of test is performed to evaluate the ILS under dynamic conditions, i.e. by considering the RFID Reader set on the body of a rescue worker moving in the area. Various configurations are assumed for the Reader positioning, such as fire-

4 fighter s helmet, shoulder, chest. As shown in Fig. 5, we consider three different movement trajectories of the first responder with respect to the Tag: radial, tangent, secant. In addition, for each trajectory we carry out tests in two directions: towards and away from the Tag. Moreover, in order to simulate different operative conditions, we repeat each test by varying the speed of the movement. In particular, we assume two speed levels: amble, equal to about 0,75 m/s, and fast, circa 1,5 m/s. Since the aim is to assess the capability of the RFID system to localize first responders, all the tests of this class are carried out by measuring the count parameter. Indeed, the number of tag detections during the movement time of each test is directly related to the identification of the rescuer. For the purpose of this test class, the following tests are carried out: Test II.1: reader on the firefighter s chest and tags on the walls of the same side of the pectoral; Test II.2: reader on the firefighter s chest and tags on the walls of the opposite side of the pectoral; Test II.3: reader on the firefighter s shoulder and tags on the walls of the same side of the shoulder; Test II.4: reader on the firefighter s shoulder and tags on the walls of the opposite side of the shoulder; Test II.5: reader on the firefighter s helmet and tags on the walls; Test III.3: presence of metal objects. This test is carried out to evaluate the effects on the identification performance due to the presence of furniture inside the rooms. Indeed, reflective objects like metal desks and cabinets can cause missed identifications or ambiguous detections, like localization of the rescuer in rooms adjacent to the real one. VI. RESULTS We carried out several tests to evaluate the performance of the RFID-based ILS in the considered scenario. In order to collect mean values of the interest parameters, every static test (Class I and partially Class III) is averaged by collecting data for 60 seconds, while each dynamic test (Class II and partially Class III) is repeated 5 times. In Fig. 6 we show results of Test I.1 performed in different test environments. The typical on-off behavior of the RFID system is observed. In particular, based on the considered test environment we obtain the maximum system efficiency for distances ranging from about 3 to 4 meters. For slightly greater distances, the efficiency rapidly decreases up to reach the complete inefficiency of the ILS. As regards the tests to characterize the radiation pattern of Fig. 5. Setup configuration for Class II Tests. System efficiency vs operative distance for different indoor environ- Fig. 6. ments. III) Class III - Presence of obstacles: this type of test is carried out to assess the system performance in presence of obstacles. In particular, we aim at verifying the limits of the system when NLOS conditions occur due various types of obstructions, such as walls, doors, tables, building floors, etc. Body losses are also evaluated. This test class include the following tests: Test III.1: characterization of walls. This test is performed to verify the capability of the ILS to avoid the typical scenario of ambiguous detection of the first responder within a narrow corridor, whereas he/she is inside a room adjacent to the corridor. Test III.2: discerning floors. This test is aimed at verifying the robustness of the ILS against the case of localization of the rescuer at the lower floor when he/she is upstairs. both the Reader and the Tag antennas, results show that the system efficiency rapidly decreases when both the distance between RFID devices and their elevation angles increase. As an example, from Test I.2 we derive that for an elevation angle equal to 30 the distance must be decreased from 2.7 m to about 1.2 m in order to keep η = 1. In Fig. 7 the system efficiency is reported as a function of the azimuth angle of the RFID Tag. For distances between the Tag and the Reader below 1.2 meters, we always measured η = 1 even for θ = 0/180. The angle of maximum system efficiency decreases with the increase of the distance, up to reach about 40 for d=3 m. In Table I, we show results of tests II.1, II.3 and II.5 for radial trajectories. The first value indicates movement direction towards the Tag, whereas the second value is related to the opposite direction. When the movement direction is towards the Tag, the Reader position on the helmet provide the best performance due to the high visibility without body obstructions.

5 d [m] Count (Test II.1) Count (Test II.2) 1 26/12 22/ /13 17/6 2 15/12 2/ /8 5/2 3 21/11 4/ /8 3/3 4 17/10 4/ /11 2/2 5 8/5 1/1 TABLE II. Count VALUES OF TEST II.1 AND TEST II.2 FOR amble/fast TANGENT TRAJECTORIES AT DIFFERENT DISTANCES d FROM THE TAG. Fig. 7. Polar representation of the system efficiency as a function of the azimuth angle of the RFID Tag for different distances between Tag and Reader. The chest positioning provides better performance than the shoulder case, but when the first responder moves away from the Tag, the count parameter is lower. This is due to the slight upward slope of the Reader when positioned on the shoulder, that improves the electromagnetic visibility between Tag and Reader in respect to the case of the Reader on the chest, in which the body quickly hinders communications between the RFID devices. In respect to an amble movement, a very sligth reduction of the count is measured when a fast speed is assumed. In Tables II and III we show results of tests II.1 and II.2 for TABLE I. Speed Reader position Count chest 19/10 Amble shoulder 15/18 helmet 32/9 chest 17/6 Fast shoulder 12/15 helmet 31/7 Count VALUES OF TESTS II.1, II.3 AND II.5 FOR RADIAL TRAJECTORIES towards/away from THE TAG. tangent and secant trajectories, respectively. As in Table I, the two reported values indicate movement direction towards and away from the Tag, respectively. Obtained outcomes show a good resilience of the system even when the Tag is on the wall at the opposite side with respect to the Reader. Indeed, the count is not null even for distances above 4 meters. We measured a quite regular trend of the count with respect to the distance increase, except for the case of 2.5 meters, where a performance reduction is registered, probably due to multi-path effects. As explained, this class of test is aimed at assessing the system performance when obstacles are present in the environment. In order to evaluate the capability of the ILS to discriminate different ambients/rooms, we carried out Test III.1 in various scenarios. We consider different wall types, i.e., light walls (wooden or light brick), heavy bricks (typical of historical buildings) and concrete with steel reinforcement. Obtained results indicate that, based on the considered environment, the first type of wall can cause high error probability of the rescuer positions within building areas (rooms, corridors, etc.). Indeed, such wall material does not significantly attenuate the signal propagation and therefore the Reader may identify a Tag positioned in adjacent rooms. Actually, in many cases the maximum operative distance of the system does not show marked reductions compared to that obtained in ideal conditions. On the contrary, the other two wall types do not allow the signal crossing, thus ensuring the absence of errors caused by ambiguous detections. As for the Test III.2, we installed the Tag on the wall adjacent to the ceiling, while the Reader is upstairs. We considered different Tag inclinations with respect to the parallel to the ceiling (uptilt), ranging from 0 (Tag on the wall) up to 60 (Tag pointing to the ceiling). Note that the uptilt configuration is not of interest for the rescue scenario. However, we assume such configurations to stress the behaviour of the ILS. When the Reader is not in proximity of the floor (> 5cm), the measured count parameter is null for all the considered uptilt values, showing the capability of the ILS to avoid possible floor ambiguity. Test III.3 is carried out to examine the impact on the ILS performance of metal object present in the test environment. The Tag is installed on the wall of a room with metal desks and cabinets. It is positioned at a height of 2.3 meters with a downtilt of 30. The Reader is positioned on the chest of the rescuer moving at low speed (amble) around the room according to four different paths, indicated by the blu lines in Fig. 8, where the red point represents the Tag. In Fig. 8 we also report the obtained results in terms of the count parameter. Test results show that the proposed RFID-based ILS can correctly discriminate the first responders positions inside restricted indoor environments like rooms and corridors, in accordance with the scenario requirements. Speed d [m] Count (Test II.1) Count (Test II.2) 1 20/10 15/ /12 10/3 2 20/19 11/ /13 8/0 Amble 3 11/10 9/ /10 12/0 4 8/13 7/ /12 6/0 5 3/12 0/0 1 12/4 9/ /4 8/0 2 10/7 5/ /7 4/0 Fast 3 7/3 3/ /2 5/0 4 7/2 6/ /5 5/0 5 5/4 3/1 TABLE III. Count VALUES OF TEST II.1 AND TEST II.2 FOR SECANT TRAJECTORIES AT DIFFERENT DISTANCES d towards/away from THE TAG.

6 Fig. 8. Paths in the room for Test III.3. VII. CONCLUSIONS In this work, we experimentally characterized the performance of a RFID-based ILS for first responders in emergency scenarios. We first defined the system architecture and we designed the experimental setup. We carried out various types of test to evaluate the capability of the ILS to identify first responders in different conditions. We considered (i.) ideal conditions, i.e. LOS transmissions, regular environmental setting, etc., (ii.) typical configuration, i.e., a typical Tag and Reader installations and dynamic conditions, and (iii.) the presence of obstacles, such as walls and metal objects. Obtained outcomes can be useful to properly set ILS deployment metrics such as number of tags and their minimum inter-distance, installation angles of tags, reader positioning on the rescuer, etc. Test results show that, by means of a correct dimensioning of the ILS, in all considered cases high reliability and accuracy can be achieved in rescuers localization. ACKNOWLEDGEMENT This work has been partially performed within the research project REference implementation of interoperable indoor location & communication systems for FIrst REsponders (RE- FIRE) co-funded by the European Commission. REFERENCES [1] H. Liu, H. Darabi, P. Banerjee, J. Liu, Survey of Wireless Indoor Positioning Techniques and Systems, IEEE Trans. Sys., Man and Cyb. - Part C: Appl. and Rev., Vol. 37, No. 6, Nov [2] R.M. Alkan, H. Karaman, M. Sahin, GPS, GALILEO and GLONASS satellite navigation systems & GPS modernization, Int. Conf. Rec. Adv. Space Tech. (RAST 2005), 2005, p [3] J. A. M. Ladd et al., On the Feasibility of Using Wireless Ethernet for Indoor Localization, IEEE Trans. Wir. Com., 5 (10), 2006, p [4] The National Institute for Occupational Safety and Health (NIOSH), Fire Fighter Fatality Investigation and Prevention Program. Available: [5] Calabrese A., Costa R., Menichini T., Rosati F., Sanfelice, G. (2013) Turning corporate social responsibility-driven opportunities in competitive advantages: A two-dimensional model, Knowledge and Process Management, 20(1), pp [6] Costa, R., & Menichini, T. (2013). A multidimensional approach for CSR assessment: the importance of the stakeholder perception. Expert Systems With Applications, 40(1), [7] Capece G., Costa R. (2009). Measuring knowledge creation in Virtual Teams through the Social Network Analysis, Knowledge Management Research & Practice, 7(4), pp [8] Costa, R., Evangelista, S. (2008). An AHP approach to assess brand intangible assets, Measuring Business Excellence,12 (2), pp [9] M. Petracca, M. Vari, F. Vatalaro, G. Lubello, Performance evaluation of GSM robustness against smart jamming attacks, 5th International Symposium on Communications Control and Signal Processing, ISCCSP [10] F. Mazzenga, M. Petracca, R. Pomposini, F. Vatalaro, R. Giuliano, Algorithms for dynamic frequency selection for femto-cells of different operators, IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2010, pp [11] M. Petracca, R. Giuliano, F. Mazzenga, Application of UWB technology for underlay signaling in cognitive radio networks, 2012, Recent Patents on Computer Science 5 (2), pp [12] F. Facchini, R. Giuliano, F. Mazzenga, Ultra-wideband detect and avoid procedure for WiMAX victims, 2009, IET Communications 3 (2), pp [13] R. Giuliano, F. Mazzenga, Capacity analysis for UWB systems with power controlled terminals under power and coexistence constraints, 2006, IEEE Transactions on Wireless Communications 5 (11), pp [14] R. Giuliano, G. Guidoni, F. Mazzenga, F. Vatalaro, On the UWB coexistence with UMTS terminals, IEEE International Conference on Communications (ICC 2004) 6, pp [15] Mazzenga, F., Simonetta, A., Giuliano, R., Vari, M., Applications of smart tagged RFID tapes for localization services in historical and cultural heritage environments, Workshop on Enabling Technologies: Infrastructure for Collaborative Enterprises, WETICE 2010, pp [16] De Luca, D., Mazzenga, F., Monti, C., Vari, M., Performance evaluation of indoor localization techniques based on RF power measurements from active or passive devices, Eurasip Journal on Applied Signal Processing 2006, pp [17] D. Boontrai, Indoor localization technique using passive RFID tags, 9th Int. Sym. on Comm. and Inf. Tech., pp , Sept [18] A. Athalye et al., A Radio Frequency Identification System for accurate indoor localization, IEEE Int. Conf. Acoust., Speech Sign. Proc., p , May [19] J. R. Guerrieri et al., RFID-Assisted Indoor Localization and Communication for First Responders 1st European Conf. on Ant. and Prop. (EuCAP 2006), pp. 1-6, Nov [20] A. R. Jimenez Ruiz et al., Accurate Pedestrian Indoor Navigation by Tightly Coupling Foot-Mounted IMU and RFID Measurements, IEEE Tran. Instr. Meas., Vol. 61, No. 1, Jan [21] R. Giuliano, F. Mazzenga, M. Petracca, R. Pomposini, Performance evaluation of an opportunistic distributed power control procedure for wireless multiple access, 5th International Symposium on Communications Control and Signal Processing, ISCCSP [22] Giuliano R., Mazzenga F., Petracca M., Consumed Power Analysis for Mobile Radio System Dimensioning, IEEE International Conference on Communications (ICC 2013), Budapest, Hungary, Jun [23] R. Giuliano, F. Mazzenga, Exponential effective SINR approximations for OFDM/OFDMA-based cellular system planning, 2009, IEEE Transactions on Wireless Communications 8 (9), pp [24] Durantini, A., Petracca, M., Ananasso, F., Experimental evaluation of IEEE WiMAX performances at 2.5 GHz band, International Wireless Communications and Mobile Computing Conference, IWCMC 2008, pp [25] Durantini, A., Petracca, M., Vatalaro, F., Civardi, A., Ananasso, F., Integration of broadband wireless technologies and PMR systems for professional communications, 4th International Conference on Networking and Services, ICNS 2008, pp [26] Omni-ID Ultra Long Range RFID Tag datasheet. Available: DataSheet.pdf [27] CAEN A528 OEM UHF multiregional compact reader datasheet. Available: [28] CAEN EASY2READ EASY CONTROLLER software. Available: