Wireless Sensor Network for Mobile Entities ...

1 downloads 0 Views 418KB Size Report
computation and sensing areas have been stimulating ... successfully used in a wide range of markets and is expected to play ... model. A wide variety of path loss models have been ..... flops of the N-hop multilateration primitive for node.
Wireless Sensor Network for Mobile Entities Localization People Monitor António Pereira, Arnaldo Monteiro, Lino Nunes, and Nuno Costa Superior School of Technology and Management Polytechnic Institute of Leiria Morro do Lena – Alto do Vieiro, 2411-901 Leiria, Portugal {eic09028, eic09047}@student.estg.ipleiria.pt {apereira, nuno.costa}@estg.ipleiria.pt Abstract The new developments in the communication, computation and sensing areas have been stimulating the hardware components miniaturization and optimization for the past few years. This evolution provided the ascension of wireless sensor networks (WSNs). This paper reports on our experience with the implementation and deployment of a Wireless Sensor Network capable of monitoring mobile entities in a specific area. This project included the development of a network management application called “People Monitor” and the network nodes application in order to detect the presence and absence of entities, in this case, human beings.

1. Introduction WSNs have an endless array of potential applications in both military and civilian applications, including robotic land-mine detection, battlefield surveillance, target tracking, environmental monitoring, wildfire detection, and traffic regulation, to name just a few. One common feature shared by all of these critical applications is the vitality of sensor location. Applications for indoor location and tracking systems can be placed in three main categories; commercial, public safety and military applications [1]. In commercial applications the need for locating patients in a hospital, guiding blind people or tracking small children or elderly individuals is of great importance as well as locating specific and important objects in warehouses or in-demand objects in hospitals. Public safety application includes locating inmates in a prison or fire-fighters in a burning building. In military applications the main interest is locating soldiers in combat. Our work consists in a

surveillance system for locate and track mobile entities in an indoor environment. This application can be deployed to track children’s, animals and also mobile assets.

2. Motivation Recently indoor location and tracking applications have attracted considerable attention in the field of telecommunication. In certain applications, the users have an RF tag that can be worn and while walking through a building they can be located accurately. This could be implemented in schools where the youngsters could be tagged so that the teachers knows exactly where they are at all times. In addition this technology can be used in hospitals to locate patients or in-demand equipment and medications. The harsh site-specific multipath environment in indoor areas introduces difficulties in accurately tracking the position of objects or people. The growing interest and demand for such applications dictates examining position estimation more carefully. The indoor channel poses a serious challenge to system designers due to the harsh multipath environment. The behaviour of the channel changes from building to building and even within a single floor of a building. The channel may vary with added objects and people moving in the vicinity. As a result, considerable work is needed for modelling the indoor channel for location and tracking applications.

3. Related work Several schemes, broadly classified into two categories, have been proposed for dealing with the localization. First, the range-based schemes need either node-to-node distances or angles for estimating locations [2]–[9]. The information can be obtained

using time of arrival (TOA), time difference of arrival (TDOA), angle of arrival (AOA), and received signal strength indicator (RSSI) methods [10]–[14]. The range-based schemes typically have higher location accuracy but require additional hardware to measure distances or angles. Second, the range-free schemes do not need the distance or angle information for localization [15]–[19]. Evennou, Marx and Nacivet described a 2D experimental method based on time differences of arrival (TDOA) using Ultra wideband (UWB) signals [20] that improve the accuracy in multipath environments. The large bandwidth provides high time-domain resolution which in return provides better ranging accuracy. Another relatively new technology used for identifying and tracking objects within a few square-meters is Radio Frequency Identification (RF-ID). The advantages of the wireless RF-ID system include identification at a distance, hands-free operation, versatile memory and processing requirements, and high accuracy due to the very short operating range [21]. RF-ID technology has been successfully used in a wide range of markets and is expected to play primary role in future mobile location applications since it enables the automated data collection and tracking of objects as they move across a limited geographical area. In RSS-based indoor location and tracking, as it is obvious from its name, the received signal strength is measured at the receiver. The RSS is related to the distance between the transmitter and the receiver mathematically in the form of path loss models [22]. The path loss models portray the signal power attenuation as the signal travels through the indoor environment. If the path loss model is known in advance then the distance between the transmitter and receiver can be calculated by measuring the received signal strength and comparing to the known path loss model. A wide variety of path loss models have been developed for different environments, each with different values of model parameters or different parameters and mathematical function forms [23].

The Mica mote (Mica) is responsible for communication, most of the processing, and most of the sensor interfaces. The Mica is a commercially available platform originally built for large scale distributed networks [24]. It is based on the ATmega 128L processor, a low-power AVR 8-bit processor with 128Kbytes of flash program memory, 4Kbytes of EEPROM and 4 Kbytes internal SRAM. The ATmega128L also includes an 8-channel 10-bit ADC, three timers, and several bus interfaces including SPI, I2C for Mica2 and DIO for Mica2 and Mica2Dot. One of the unique features of the Mica mote is the integrated 916MHz RFM radio used for communication. Maximum speed is currently set to 38.4Kbps and ranges from 50m indoor NLOS to 175m outdoor LOS. The radio may also be used to program the Micas. When using more than ten Micas for an application, network programming is essential for the sanity of the programmer. An image of the Mica2 and Mica2Dot is shown in Figure 1.

Figure 1. Mica2 and Mica2Dot motes

4.2. MIB510 serial gateway The MIB510CA allows for the aggregation of sensor network data on a PC as well as other standard computer platforms. Any MICA2/MICA2Dot node can function as a base station when embedded in the MIB510CA serial interface board. In addition to data transfer, the MIB510CA also provides an RS-232 serial programming interface. An image of the MIB510CA is shown in Figure 2.

4. Hardware architecture The PeopleMonitor wireless sensor network is built entirely from off-the-shelf components with the goal of remaining as simple and flexible as possible. Two main components comprise the standard PeopleMonitor hardware configuration: Mica Mote and a network base station board.

For this first project prototype, the implemented network topology is the single-hop, but in the future multi-hop will be implemented in order to more easily survey the mobile entity localization.

4.1. Mica mote

5. Software architecture

Figure 2. MIB510 Serial Gateway

Software is based on the event-driven TinyOS operating system developed at UC Berkeley for sensor networks [25]. It has a model based on components programmed by the language nesC, similar to C. The operative system is capable to accomplish multiple tasks and events required by the wireless network sensors with a minimum hardware requirement, such as memory, processor and energy consumption. TinyOS offers several other advantages as well, including the ability to abstract software modules. TinyOS is an open-source, component-based software platform. This is a great advantage in the reuse of software. Many components, such as the network stack and lower level driver components are already written. TinyOS software is available in the SourceForge Repository [26]. The link layer protocol used is the B-MAC that is available in the TinyOS software. Software for the PeopleMonitor wireless sensor network is divided into three parts: software that runs on the mobile entity, software that runs on the static entity and software that runs on the gateway.

People Monitor application is divided in the following logical parts: Graphical interface or GUI, Analysis and Storage and TCP Client. The graphical interface concerns on forms and graphical methods that interact with the user. The Analysis and Storage is an intermediate layer between the graphical interface and the wireless sensor network and has mechanisms to register and to configure the sensors data. It also calculates the mobile sensors localization in the existing sensed areas. The network connection is established through the Serial Forwarder TCP Server that receives the sensor’s data from a serial port that is connected to the gateway.

5.1. Static entity The static entity is responsible for listening the mobile sensors and gateway messages in the network. When a messages is received from a mobile entity it attaches its identification and its received signal strength indicator (RSSI), forwarding it to the gateway. The gateway messages are forwarded in a broadcast mode to the static entity

Figure 3. People monitor GUI The network connection is established through the Serial Forwarder TCP Server that receives the sensor’s data from a serial port that is connected to the gateway.

5.2. Mobile entity The mobile sensor application periodically sends a message into the wireless sensor network, with its identification. Also the mobile entity receives commands forwarded by gateway in order to change values, such as, the sending messages interval time and the transmitted power. This entity will be able to modify more parameters, in the future.

5.3. Gateway This entity is only responsible for receiving messages forwarded from the interface radio into the serial interface and vice-versa. It has a buffering messages mechanism. The gateway application is part of the TinyOS distribution [25].

6. People monitor application

Figure 4. High level architecture

7. Application protocol The applicational protocol was based on an existing one in the TinyOS operating system. The exchanged messages between the mobile entity and the static entity and, between the static entity and the gateway are demonstrated in the following figure.

approaches and try to improve or tune them in order to achieve better energy savings.

8. Analysis and current results

Figure 5. Mobile and static message structures The messages proceeding from the mobile entity are forwarded into the gateway through the static entity.

Figure 6. Message flow All the sensors in the network periodically send a message to notify the battery voltage. The following figure represents the command messages sent between the PeopleMonitor application and the mobile entities.

Analyzing this monitoring method for mobile entities localization and tracking concludes that there is a minimum margin of error. This method allows the mobile entities location in areas like classroom and offices. Depending on the amount of static entities in the area, it is possible to reduce the area where the mobile entity is. Due to the fact that the network topology is single-hop, it is not possible at this stage to cover a significant area. The most important tests were: the PeopleMonitor application detects the mobile entity in the area that it is physically present at any moment and PeopleMonitor application detects the mobile entity transition between areas. One evidenced in the first test is that the mobile entity was always detected in the area that it was physically present. In the second test a little time shift in the update of the transition between areas were verified. This phenomenon happened, first due to loss of packages by the static entity and second because the radio signal reflections on the walls and objects. Some simple characterization experiments have been used to measure the PeopleMonitor performance in a typical lab setting. The mobile entity was moving along the corridor and was taken samples of the RSSI values.

Figure 8. Received power in dBm versus Distance (indoor measurement LOS, 30cm above ground) This test was useful for the characterization of the RF channel in this particular case - indoor environment. This denotes that the propagation of the RF signals is not linear. Figure 7. Command message flow In this first prototype we used the push and pull approaches. On future work we will review these

9. Lessons learned As mentioned earlier there are different metrics that can be used in indoor positioning. Although this research focuses on RSS-based indoor location it is worth mentioning the other techniques. In AOA-based

indoor location direction-based triangulation is used, where two or more reference points are used to determine the position of the mobile entity. The AOA is usually measured with directional antennas or antenna arrays. This metric is not preferable in indoor environment because of the harsh multipath which introduces inaccuracies into the detection of the AOA in both LOS and NLOS conditions. TOA is another metric which have been used in many positioning systems. Since the speed of the signal is known the time that first path arrives in receiver is directly related to the distance between mobile terminal and fixed terminal. Triangulation is used to determine the exact location of the mobile terminal considering the time that first path has arrived. For this purpose the transmitter and receiver should be synchronous which is costly and needs more complicated circuitry both in transmitter and receiver. TDOA technique uses the time difference of propagation in signals to calculate the exact location of the mobile entity. Systems like Cricket [27] uses TDOA between RF and acoustic pulses to estimate within a few centimeters the exact position of the mobile entity, because speed of acoustic waves in air is slow compared to that of RF. In RSS-based indoor location, as it is obvious from its name the received signal strength is measured at the receiver. The RSS is related to the distance between the transmitter and the receiver mathematically in the form of path loss models [28]. The path loss models portray the signal power attenuation as the signal travels through the indoor environment. If the path loss model is known in advance then the distance between the transmitter and receiver can be calculated by measuring the received signal strength and comparing to the known path loss model. A wide variety of path loss models have been developed for different environments, each with different values of model parameters or different parameters and mathematical function forms [29]. Besides using metrics directly, there is another method to determine the location of the mobile terminal, fingerprinting. Fingerprinting basically means building database with respect to the received signal’s metric related to the present location of the mobile terminal. Operation of such systems has 2 phases, named off-line phase and on-line phase. Off-line phase is the data collection phase or learning phase. The major portion of this phase involves roaming around in the site and collecting data. Recording the set of information as a function of the user’s location covering the entire zone of interest, we can form a database consisting of fingerprints. Each fingerprint corresponds to fingerprint information associated with a known user’s location.

Real-time phase or on-line phase is the phase of matching the obtained fingerprint for the specific location and compared with the database. A pattern matching algorithm is then used to identify the recorded information closest to the new specific fingerprint. K. Lorincz and M. Welsh described a robust, decentralized approach to RF-based location tracking, using fingerprinting [30]. Any change in the environment (moving people, furniture rearrangement) alters the signal strength and impinges on the accuracy and reliability of fingerprinting based solutions. This solution implements a small potentiality part of this technology. It has very much to improve, like the introduction of security mechanisms in the messages exchanged between the entities, tolerance to failures, energy efficiency and hot configurability in some levels.

10. Conclusions and future work This project allowed the development of an application that monitors mobile objects in one determined area using RSS-based indoor location. The implementation of this system with the current technology, hardware and software turned to be efficient when used in small areas. The study and experiences in larger areas are necessary to verify its scalability and reliability, since there is growing community that is very interested on this technology development. Also the investigation of new techniques for localization and tracking are mandatory to improve the accuracy is our future direction.

11. References [1] M. Heidari, A Testbed for Real-Time Performance Evaluation of RSS-based Indoor Geolocation Systems in Laboratory Environment, Faculty of Worcester Polytechnic Institute, 2005 [2] J. Hightower,G. Boriello, and R.Want, “SpotON: An indoor 3D location sensing technology based on RF signal strength,” Univ. of Washington, Tech. Rep. UW CSE 00-0202, Feb. 2000. [3] P. Bahl and V. N. Padmanabhan, “RADAR: an inbuilding RF-based user location and tracking system,” in Proc. IEEE Joint Conf. IEEE Computer Communications Societies (INFOCOM), Tel Aviv, Israel, Mar. 2000, pp.775– 784. [4] P. Bergamo and G. Mazzini, “Localization in sensor networks with fading and mobility,” in Proc. IEEE Int. Symp. Personal, Indoor Mobile Radio Communications (PIMRC), Lisbon, Portugal, Sep. 2002, pp.750–754.

[5] N. B. Priyantha, A. Chakraborty, and H. Balakrishnan, “The cricket location-support system,” in Proc. ACMInt. Conf. Mobile Computing Networking (MOBICOM), Boston, MA, Aug. 2000, pp. 32–43.

[18] C. Savarese, J. Rabaey, and K. Langendoen, “Robust positioning algorithms for distributed ad-hoc wireless sensor networks,” in Proc. USENIX Technical Annu. Conf.,Monterey, CA, June 2002, pp. 317–327.

[6] A. Savvides, C. C. Han, and M. B. Srivastava, “Dynamic fine-grained localization in ad-hoc networks of sensors,” in Proc. ACM Int. Conf. Mobile Computing Networking (MOBICOM), Rome, Italy, July 2001, pp. 166–179.

[19] T. He, C. Huang, B. M. Blum, J. A. Stankovic, and T. Abdelzaher, “Range-free localization schemes for large scale sensor networks,” in Proc. ACM Int. Conf. Mobile Computing Networking (MOBICOM), San Diego, CA, Sep. 2003, pp. 81–95.

[7] A. Savvides, H. Park, and M. Srivastava, “The bits and flops of the N-hop multilateration primitive for node localization problems,” in Proc. ACM Int. Workshop Wireless Sensor Networks and Applications (WSNA), Atlanta, GA, Sep. 2002, pp. 112–121. [8] D. Niculescu and B. Nath, “Ad Hoc Positioning System (APS) using AoA,” in Proc. IEEE Joint Conf. IEEE Computer Communications Societies (INFOCOM), San Francisco, CA, USA, Mar. 2003, pp. 1734–1743. [9] A. Nasipuri and K. Li, “Adirectionality based location discovery scheme for wireless sensor networks,” in Proc. ACM Int. Workshop Wireless Sensor Networks Applications (WSNA), Atlanta, GA, Sep. 2002, pp. 105–111. [10] T. S. Rappaport, J. H. Reed, and B. D.Woerner, “Position location using wireless communications on highways of the future,” IEEE Commun. Mag., vol. 34, pp. 33–42, Oct. 1996.

[20] F. Evennou, F. Marx and S. Nacivet, “An Experimental TDOA UWB Location System for NLOS Environments,” Division R&D, TECH/ONE. [21] A. Brewer, N. Sloan, and T.L. Landers, “Intelligent Tracking in Manufacturing”, Journal of Intelligent Manufacturing, Vol. 10, 1999. [22] K. Pahlavan, X, Li, J. P. Makela, “Indoor Geolocation Science and Technology” IEEE communication Mag. Vol 40, no 2, Feb. 2002, pp 112-118 [23] K. Pahlavan and A. Levesque, “Wireless Information Networks,” John Wiley & Sons, 1995 [24] http://www.xbow.com [25] http://www.tinyos.net [26] http://sourceforge.net/projects/tinyos/

[11] J. Caffery Jr. and G. L. Stüer, “Subscriber location in CDMA cellular networks,” IEEE Trans. Veh. Technol., vol. 47, pp. 406–416, May 1998. [12] R. Klukas and M. Fattouche, “Line-of-sight angle of arrival estimation in the outdoor multipath environment,” IEEE Trans. Veh. Technol., vol. 47, pp. 342–351, Feb. 1998. [13] L. Cong and W. Zhuang, “Hybrid TDOAJAOA mobile user location for wideband CDMA cellular systems,” IEEE Trans. Wireless Commun., vol. 1, pp. 439–447, Jul. 2002. [14] M. McGuire, K. N. Plataniotis, and A. N. Venetsanopoulos, “Location of mobile terminals using time measurements and survey points,” IEEE Trans. Veh. Technol., vol. 52, pp. 999–1011, Jul. 2003. [15] N. Bulusu, J. Heidemann, and D. Estrin, “GPS-less low cost outdoor localization for very small devices,” IEEE Personal Commun., vol. 7, pp. 28–34, Oct. 2000. [16] N. Bulusu, J. Heidemann, D. Estrin, “Adaptive beacon placement,” In Proc. IEEE Int. Conf. Distributed Computing Systems (ICDCS), Phoenix, AZ, Apr. 2001, pp. 489–498. [17] D. Niculescu and B. Nath, “DV based positioning in ad hoc networks,” Kluwer J. Telecommun. Syst., vol. 22, no. 1, pp. 267–280, Jan. 2003.

[27] H. Balakrishnan, R. Baliga, D. Curtis, M. Goraczko, A. Miu, N. B. Priyantha, A. Smith, K. Steele, S. Teller, K. Wang, “Lessons from Developing and Deploying the Cricket Indoor Location System,” November 2003 [28] K. Pahlavan, X, Li, J. P. Makela, “Indoor Geolocation Science and Technology” IEEE communication Mag. Vol 40, no 2, Feb. 2002, pp 112-118 [29] K. Pahlavan and A. Levesque, “Wireless Information Networks,” John Wiley & Sons, 1995 [30] K. Lorincz, M. Welsh, “MoteTrack: A Robust, Decentralized Approach to RF-Based Location Tracking,” In Proceedings of the International Workshop on Location and Context-Awareness (LoCA 2005) at Pervasive 2005, May 2005