This report studies various technologies which can be used for people tracking and crowd management. During any pilgrimage session, the events like losing a group member, medical emergencies, and stampede are very common. In case of disasters like flood, fire, earthquake pilgrims are often left stranded as they have a little knowledge about that area. To address these concerns, the authorities need information about the location and direction of movement of the pilgrims. Sometime, it may be necessary to identify a person in cases like a lost child, unconscious or dead person. Researchers have proposed a number of solutions using the wireless technology which includes RFID tags, NFC, GPS, Bluetooth and Wi-Fi. An introduction about the technologies and how they can be used to design systems for people identification and tracking is presented. The various systems using a set of technologies have been studied in terms of their effectiveness and the cost.

1. A seminar report
on
A Study of Wireless Technology Based Pilgrim Tracking Systems
Submitted in partial fulfilment of the requirements for the award of course credits for
INTEGRATED DUAL DEGREE
in
COMPUTER SCIENCE AND ENGINEERING
(with specialization in Information Technology)
Submitted by
DEEPAK KUMAR
IDD-CSE
(Enrollment No: 10211009)
Under the guidance of
Dr. Dhaval Patel
Assistant Professor
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
INDIAN INSTITUTE OF TECHNOLOGY
ROORKEE – 247667
August, 2014

2. ii
ABSTRACT
This report studies various technologies which can be used for people tracking and crowd management. During any pilgrimage session, the events like losing a group member, medical emergencies, and stampede are very common. In case of disasters like flood, fire, earthquake pilgrims are often left stranded as they have a little knowledge about that area. To address these concerns, the authorities need information about the location and direction of movement of the pilgrims. Sometime, it may be necessary to identify a person in cases like a lost child, unconscious or dead person. Researchers have proposed a number of solutions using the wireless technology which includes RFID tags, NFC, GPS, Bluetooth and Wi-Fi. An introduction about the technologies and how they can be used to design systems for people identification and tracking is presented. The various systems using a set of technologies have been studied in terms of their effectiveness and the cost.

3. iii
TABLE OF CONTENTS
1. INTRODUCTION ............................................................................................................................. 1
1.1 Motivation .................................................................................................................................... 1
1.2 Background .................................................................................................................................. 2
2. WIRELESS TECHNOLOGIES......................................................................................................... 3
2.1 Radio Frequency Identification (RFID) ....................................................................................... 3
2.1.1 Categorising RFID tags......................................................................................................... 3
2.1.2 Effects of operating frequency on RFID tags ....................................................................... 4
2.1.3 Ultra-Wideband RFID tags ................................................................................................... 5
2.2 Near field communication (NFC) ................................................................................................ 5
2.3 Wi-Fi ............................................................................................................................................ 6
2.3.1 Passive Wi-Fi Tracking......................................................................................................... 6
2.3.2 Active Wi-Fi tracking ........................................................................................................... 7
2.4 Bluetooth ...................................................................................................................................... 8
2.4.1 Inquiry or scan based tracking .............................................................................................. 9
2.4.2 Inquiry-free tracking ............................................................................................................. 9
2.5 GPS ............................................................................................................................................ 10
3. SYSTEM DESIGNS ........................................................................................................................ 11
3.1 Integrated Mobile and RFID System ......................................................................................... 11
3.2 Smartphone GPS ........................................................................................................................ 12
3.2.1 Updating Location .............................................................................................................. 12
3.3 Wireless Sensor Network (WSN) .............................................................................................. 13
3.4 Bluetooth .................................................................................................................................... 14
3.5 Wi-Fi .......................................................................................................................................... 15
4. SUMMARY ..................................................................................................................................... 16
5. CONCLUSION ................................................................................................................................ 18
6. REFERENCES ................................................................................................................................ 19

4. iv
LIST OF TABLES
Table 2.1: Ranges and costs of RFID tags ............................................................................................. 5
Table 4.1: Comparison of wireless technologies ................................................................................. 16
LIST OF FIGURES
Figure 2.1: Waterproof RFID Wristband ............................................................................................... 3
Figure 2.2: Process of associating with an access point ........................................................................ 8
Figure 3.1: Architecture of a system using RFID and Smartphone GPS............................................. 12
Figure 3.2: HajjLocator Architecture ................................................................................................... 13

5. 1
1. INTRODUCTION
1.1 Motivation
Pilgrimage has a great significance in India. Each pilgrimage session attracts a huge crowd. Usually the pilgrims move simultaneously in a large group. Getting lost in the crowd is very common. Finding the lost person among thousands of pilgrims causes a lot of trouble for his relatives and the helping teams. As the pilgrims come from across the country, communication problems may also arise due to lack of a common language, which makes it hard to get the help from the helping teams.
Overcrowding at a specific site often results in stampede which leaves many people dead and injured. Sometimes the dead persons cannot be identified. The medical history of an injured person can provide better treatment and his family members should be notified as soon as possible. We can get the home address and medical history about a person only after that person is identified and we have that information already stored.
The problem of identifying and locating a pilgrim can be solved if a device can be associated with the pilgrim. The device can be monitored to get the location trajectory of the pilgrim. Having the trajectories of the movements of all persons, the crowd management can be made easy. It can help in preparing traffic plans. We can get the statistics about the most visited places, the duration of staying at a site and the way pilgrims disperse. The number of pilgrims at a specific site can be restricted to a safe limit. The group in which a person is moving can be identified automatically and the lost situations can be predicted if the system detects that the distance of a person is increasing from his group.
The disasters like flood, fire, and earthquake leave the people stranded. The need may arise to quickly relocate the pilgrims. Having an estimate of the number of pilgrims, their direction and speed of movement may help in quickly relocate the pilgrims. The trajectory data can be used to find some anomalies like people going from an alternate route which may be due to broken bridge or traffic jam. Crowd gathering at unexpected place can also be detected which may be due to an accident or their path is blocked due to flood. Ambulances or other moving objects can also be tracked by using methods similar to tracking pilgrims.
In this report, a survey of various systems designed using wireless technologies for pilgrim tracking is presented. Some other systems for tracking people in general are also studied, which can be utilized for the purpose of tracking pilgrims. The technologies and system designs are studied for the features they can provide like identifying or tracking a person, the infrastructure cost, scalability and the ease of deployment.

6. 2
1.2 Background
Tracking and monitoring the pilgrims in a crowded area is a difficult task. Researchers have tried a lot of solutions using smartphones [1] [2], RFID [3], wireless sensor networks (WSN) [4] or a combination of them [5] [6]. Some of them deeply studied the protocol of Wi-Fi and Bluetooth to track the smartphones using unconventional techniques [7].
Today smartphones come equipped with GPS, Wi-Fi, Bluetooth, NFC, accelerometer, compass and other sensors. Most of the systems utilize only the GPS and the mobile network available in smartphones, using Wi-Fi, Bluetooth, NFC and other sensors like accelerometer and compass remain unused. These systems have a server maintaining the database of locations of each pilgrim and a smartphone application for the pilgrims which get the coordinates using GPS and uploads it to the server using GPRS, 3G or short message service (SMS). The mobile app can be used by a pilgrim to request help or locate the helping team and officers easily. Section 2.2 explains that the NFC technology can also be used to identify pilgrims.
RFID based systems are most suitable for the purpose of identifying a large number of pilgrims. Because RFID tags are cheaper they can be provided to each pilgrims and the tag can be read using a RFID reader, which can also provide the information about a pilgrim. But RFID readers are costly, using them to track the pilgrims requires installation of a number of readers along the path. Also the passive RFID readers have a short range, so the pilgrim must be close to reader. Active RFID tags can be read from a larger distance but they are significantly more costly than passive tags. Researchers have faced many difficulties in the system designed using RFID and they found it to be impractical with the crowd [2]. Section 2.1 provides a detailed overview of RFID technology and how it can be used in pilgrim tracking and management.
Systems based on Wi-Fi or Bluetooth require the mobile device to have Wi-Fi or Bluetooth enabled all the time. Wi-Fi monitors or Bluetooth monitors are installed similar to RFID readers in RFID based systems. When the mobile device comes near a monitor, it captures the packets transmitted from the mobile device and extracts it MAC address. The MAC address uniquely identifies a device and it can be associated with a pilgrim at the time of registration. Section 2.3 and 2.4 discuss the Wi-Fi and Bluetooth technology to monitor the pilgrims.
Tracking millions of pilgrims in real time requires a highly available, cost effective and robust system. So the technology must be selected carefully. Existing technologies are surveyed for their practicality in pilgrim management taking into account the cost and features supported. Chapter 3 discusses various systems proposed by researchers.

7. 3
2. WIRELESS TECHNOLOGIES
2.1 Radio Frequency Identification (RFID)
RFID is a technology which uses radio-frequency waves for automated identification of objects from a distance. In RFID systems an object is marked with a tag. These RFID tags are small, wireless devices that store an identification number, which is typically a pointer to database entry. An RFID reader does the work of scanning these tags, when they come in its read range. Deployment of RFID systems is increasing due to cost reduction. Now they provide longer read range and higher data transfer rate. RFID tags are also available as waterproof wristbands. Section 3.1 discusses the system designed for tracking pilgrims using RFID tags.
Figure 2.1: Waterproof RFID Wristband (Courtesy of [8])
2.1.1 Categorising RFID tags
RFID tags can be categorised depending on their source of power or the technology used to transfer the power and energy:
Based on source of power
1. Passive: Passive tags are most popular due to their very low cost starting from around ₹10 per tag. They get their energy by harvesting the RF energy, which comes from the reader when it tries to read the tag. They can operate only when a reader is trying to read.
2. Semi-Passive: They have their own power source but they can transmit information only when read by the reader. Due to having battery for power, semi-passive tags are significantly more costly than passive tags, and their size is also larger.
3. Active: They have their own power source and can transmit information automatically or when asked by a reader. Having battery provides significantly longer read range than passive tags, but makes it most costly tag. The chip inside tag consumes very less amount of power (less than 1 mW), so most of the tags have battery life or around 5 years.

8. 4
Based on power and data transfer
RFID tags can be divided in two categories based on power transfer from reader to tag and data transfer from tag to reader: [9]
1. Near-field RFID: The energy is transferred using magnetic induction. The RFID reader generates alternating magnetic field which produces an alternating voltage in the tag coil. The generated voltage is then rectified and a capacitor attached to it accumulates the charge. This charge is used to power the chip inside tag. Load modulation technique is used for transferring the tag id to the reader. Current in the tag coil produces its own magnetic field which opposes the reader’s magnetic field. This will change the current flow in reader’s coil which can be detected by the reader. The range for using magnetic induction is approximately 푐/2휋푓, so the read range decreases on increasing the frequency. This technique is used for tags operating at less than 100 MHz.
2. Far-field RFID: The tag has a dipole antenna which captures the EM waves transmitted from the dipole antenna of a reader. The energy received in the tag is in the form of alternating potential difference. The rectified voltage is applied to capacitor which stores the charge. The data transfer is done using back scattering. The tag reflects the part of received energy by changing the impedance. The tag’s ID can be encoded by changing the impedance over time. The reader receives this reflected signal, so the energy received is proportional to 1/푟4, where r is the distance between reader and the tag. So the read range of passive tags using this technology is around 3 to 6 meters. This technique is used for tags operating in the UHF band.
2.1.2 Effects of operating frequency on RFID tags
1. As we increase the frequency, the energy contained in radio frequency waves also increases, so the passive tags can harvest more energy, resulting in higher read range. Passive UHF tags can have read range more than 3 meters, while LF and HF tags have few centimetres.
2. At high frequencies radio reflective materials impede the scanning of tags. So UHF tags will be affected by beverages and metals. The human body has lot of liquid, so UHF tags may not work properly near humans.
3. At higher frequency the data transfer rate is also high. So UHF and UWB tags have high data transfer speed, so the tag id can have more number of bits and more number of tags can be read in a fixed time.
4. Tags operating in UHF are cheaper than the tags operating in high frequency or lower frequency.

9. 5
Table 2.1 compares passive and active tags operating in different categories with regard to their range and cost. Empty field indicates that the tags falling in that category is not commercially available.
Table 2.1: Ranges and costs of RFID tags (adapted from [10])
Frequency
Passive
Active
Range Cost Range Cost
125-134.3 kHz
Low Frequency (LF)
10 cm – 30 cm
$1
13.56 MHz High Frequency (HF) 10 cm – 1.5 m $5
865-867 MHz
Ultra High Frequency (UHF)
1 m – 15 m
$0.15
50 m
$20 2.45 or 5.8 GHz Microwave 3 m 30 m $25
3.1–10 GHz
Ultra Wide Band
NA
up to 200 m
$5
2.1.3 Ultra-Wideband RFID tags
This technology is relatively new. UWB tags operate in 3.1 to 10 GHz band, by transmitting low power signals on a large range of frequencies instead of transmitting a strong signal on a particular frequency. UWB tags are energy efficient and they provide a very long read range. Since they use a broad range of frequencies, the interference is very less. So, they can be used nearby liquids, metals and humans. They do not interfere with any other device because of weak signal on a particular frequency. So the UWB tags can be the best among active RFID tags for tracking pilgrims.
2.2 Near field communication (NFC)
NFC is a wireless technology for short range communications between the devices which are held together for a short time. Using NFC mobile devices can communicate with the nearby devices in both directions. The theoretical distance of working is up to 20 cm, while practically the devices should be within 5 cm [11]. NFC is based on RFID technology and the NFC standard is compatible with already deployed contactless solutions [12]. Mobile phones have embedded tags within them. The NFC technology uses magnetic induction for data transfer and powering the tag. An unpowered NFC chip or passive tag can communicate to an NFC device. NFC has slow data transfer rate with maximum theoretical rate being 424 kbps. The typical data transfer rate is from 30 to 60 kbps [11].

10. 6
NFC can be used for identification, if we already have the details about the person and the device. It can be used for tracking purpose also, if the pilgrims are required to tap their phones at the check post. The device can be identified even when the phone is switched off. A smartphone with NFC capability can read the high frequency RFID tags operating in 13.56 MHz frequency. So the pilgrims can be given low cost passive RFID tags which can be read using a smartphone and the information about the pilgrim can be accessed using Internet or stored in the tag.
NFC can be used to transfer data. When using GPS to log the location trajectory of the user, the logged data can be transferred to another device seamlessly. It requires only to touch another device, both of the devices can transfer data using Bluetooth or Wi-Fi. For data transfer using Bluetooth, first we need to activate the Bluetooth on both devices, then search for the other device and then pair. These functions may not be easily accessible in the menu system. The updated Bluetooth standard supports paring the devices via NFC. This whole process can be replaced by momentarily holding the devices close to each other. In the similar way NFC can be used to connect to a Wi-Fi hotspot or transfer the data using Wi-Fi Direct.
2.3 Wi-Fi
Wi-Fi operates in 2.4 GHz and 5 GHz frequency bands. It provides longer range and very high data transfer rate compared to Bluetooth. Research has been done on tracking smartphones without modification in their software using Wi-Fi [13] or Bluetooth [7]. Researchers have also compared Bluetooth and Wi-Fi in terms of architecture, discovery time and signal strength [14]. This section discusses how we can use Wi-Fi to track a smartphone, and the next section describes the use of Bluetooth for tracking.
Wi-Fi enabled devices transmit messages even when they are not connected to a network. For detecting available access point within range, probe request frames are transmitted periodically on every supported channel. These frames can be detected by using Wi-Fi monitoring equipment in the region of interest, which can provide coarse-grained location for the phones nearby monitoring device. Each frame transmitted consists of a MAC header, payload and frame check sequence (FCS). The MAC header contains the MAC address of the Wi-Fi device along with other fields. Following sections discuss tracking a Wi-Fi enabled phone passively and actively, in order.
2.3.1 Passive Wi-Fi Tracking
During passive tracking no frames are transmitted. A Wi-Fi monitor listens silently for the packets transmitted from the phone. As the smartphone software is not modified, the message transmission occurs on the discretion of phone, for example when the phone is searching for other Wi-Fi networks.

11. 7
These transmissions can be detected from 20 meters to 300 meters depending on the transmission power and other environmental factors. These frames contain the MAC address of sender. By associating this MAC address with a pilgrim, the location information of can be logged in the database.
The performance of the Wi-Fi monitoring system can be improved by increasing the number of frames transmitted from a device. More number of transmissions helps in getting more devices detected and their trajectories can be obtained more accurately. A phone may not get detected by Wi-Fi monitors. One of the possible reason can be that the phone is not transmitting any probe message. Also a message transmitted may not be correctly detected by the monitor due to environmental factor or long range. The next section uses active tracking approach to get more transmission from the smartphone.
2.3.2 Active Wi-Fi tracking
Active tracking involves transmissions of frames from a Wi-Fi monitors. A smartphone can be tricked to connect to a virtual hotspot or reply to a query, which involves transmission of frames and increases the number of devices detected and the accuracy of localizing the device. Following three techniques can be used to increase the number of frames transmitted.
Advertising popular access point’ SSID
At some places like airport, few companies provide free Wi-Fi. The hotspots from AT&T use “attwifi” as their SSID. Once a phone is connected to an “attwifi” hotspot, it can automatically connect to any other hotspot with “attwifi” as its SSID. To boost the number of frames received from a device, a beacon frame is transmitted with the SSID of a popular hotspot. If the phone has already been connected once to the hotspot with this SSID, the phone then sends an association request to this virtual access point.
To get even more frames transmitted from each device, a fully functional access point can be emulated. Providing Internet connectivity is not necessary. The smartphone occasionally transmits null frames to notify the access point about it power state, which can be used for tracking.
Opportunistic Access Point Emulation
A phone which has been associated to a particular SSID will transmit directed probe requests for it. If the phone recognizes a hotspot with that SSID, it will make attempts for connecting to the hotspot. The recognition process relies on the security protocol, which is not contained in the probe request. In case of open access point it gets associated completely, and we will get more frames.
For secured networks the phone will not attempt to associate if the security protocol doesn’t match with already remembered credentials. For associating using secure protocols like WPA, 4-way

12. 8
handshake is required. This 4-way handshake cannot be completed because we don’t have the required credentials for a successful handshake, and we don’t try to complete. But as shown in Figure 2.2, if the access point does not initiate the handshake process (step C), after association response (step A), the phone will transmit null packets (step B) for around 10 seconds to check if AP is still there and after that it will repeat the association attempt. These continuous null packets can provide better trajectory.
Figure 2.2: Process of associating with an access point (Courtesy of [13])
Sending RTS packets to detected phones
This method does not require association of a phone with the emulated access points. To boost the number of frames transmitted from a phone, an RTS (request to send) frame can be sent. IEEE 802.11 standard requires that if a station receives an RTS frame and if the channel is free, then it should send a CTS (clear to send) frame. An RTS frame has both receiver address (RA) as well as transmitter address (TA), but CTS frame has only the receiver address. So a CTS frame transmitted from a device does not identify it. To solve this problem, we can specify unique transmitter address (TA) in RTS frame for each phone. So the phone will transmit CTS with the unique transmitter address as its receiver address, which can be intercepted by the Wi-Fi monitor.
2.4 Bluetooth
Almost every phone comes with Bluetooth. It consumes less power and provides decent data transfer rate. Most Bluetooth devices are battery powered class 2 devices using 2.5 mW of power. Its practical

13. 9
operating range is about 10 m to 30 m, which makes is suitable for tracking the persons passing nearby a Bluetooth monitoring device. Bluetooth technology operates in 2.4 to 2.485 GHz (ISM) band, it uses Frequency Hopping Spread Spectrum technology at a nominal rate of 1600 hops per second. [15]
Connecting two Bluetooth enabled devices involves inquiry and paging, this process is called paring. Inquiry procedure is used to discover all the devices in range and get their device addresses and clocks. Paging procedure is used to establish the connection. During paging the master device transmits search packets (called pages) targeted to a slave device until master gets a reply.
In android Bluetooth can be enabled programmatically, which we can use to transfer the data when we are nearby some check post, which we can detect with the help of geo-fencing. And we can turn it off when leaving the geo-fence.
2.4.1 Inquiry or scan based tracking
To discover the nearby devices, a master device transmits a discovery packet on some predefined 32 channels. If a device is configured to be discoverable, it will respond to this packet which can be used to identify the device. It might take around 10 seconds to discover all the devices in range because the response will be transmitted after a random delay which minimizes the probability of response collisions from other devices. The scanning process also disrupts the current channels which are in use.
Using above method anybody can track mobile devices by sending discovery packets continuously. Considering that as privacy risk, nowadays Bluetooth is set in discoverable mode only for a limited time or when the user is on specific page as is the case in Windows Phone. A scan packet originating from a mobile device is not required to have source address, so it cannot be used for tracking the phone when it searches for other devices to communicate.
2.4.2 Inquiry-free tracking
The authors have described a system for tracking the Bluetooth devices without discoverability, which is connection based. It requires a Bluetooth device to be already connected to a Bluetooth monitor by paring. Paring insures that an unpaired device cannot track a Bluetooth enabled phone. The Bluetooth monitor sends an inquiry which is targeted for a particular device instead of all devices. It will require multiple inquiry messages for tracking a number of devices, but the advantage is that a specific device can be targeted quickly. Section 3.4 describes a system using Bluetooth for tracking persons.

14. 10
2.5 GPS
Global positioning system is a navigation system which provides position, velocity and time information anywhere on earth. GPS uses the signal incoming from more than 4-5 satellites to calculate its position. But to send this location to the server it requires other means of connectivity like Wi-Fi, Bluetooth, 3G etc. Accuracy of GPS is from 2 to 10 meters, but due to weather conditions or indoors it may be more than 100 meters, so the indoor use of GPS is very limited.
The Global Navigation Satellite System (GLONASS) also works similar to GPS. In most of the smartphones having Snapdragon processor, both satellite networks GPS as well as GLONASS can be accessed, which improves the location services in terms of reliability, increased accuracy and faster fix on location [16]. Standalone GPS takes time to get its first location fix which may be up to 30-40 seconds. This time can be reduce by using assisted GPS (A-GPS) which gets the almanac and ephemeris data from the Internet instead of getting it from the satellites by utilizing the cell phone tower’s id (Cell ID).
Getting real time position from a smartphone requires an application to be developed, which gets the position from GPS device, aggregates or stores the data and then transfer it to the central server when the network connectivity is available. Section 3.2 describes a system designed for tracking pilgrims using smartphone GPS.

15. 11
3. SYSTEM DESIGNS
Various systems have been designed by the researchers using the GPS of a smartphone, RFID, standalone GPS with wireless communication for tracking the pilgrims. Some systems have been designed to monitor presence of people in an area using Wi-Fi or Bluetooth, not specifically for tracking pilgrims, but they can be customized to track pilgrims.
The system should be scalable to handle real time location updates of millions of pilgrims. The system should also be inter-operable and should integrate with other systems. It must be designed by keeping in mind the duration of time it will be used. Using expensive infrastructure for an infrequent event cannot be justified.
System designed for tracking people usually incorporate location based services (LBS) and geographic information system (GIS). Location based services are services which use the location of an object or person to control the features it provides. For example: finding nearby ATM, getting route to it and traffic information, advertising, road traffic optimization, emergency call positioning. Location of a device can be obtained using a wide range of complementary technologies including GPS, Assisted GPS, RFID, cell identification, Wi-Fi positioning system. Following sections discuss the various system designs possible for tracking pilgrims.
3.1 Integrated Mobile and RFID System
This system is designed for tracking the pilgrims during Hajj. In this system, an RFID tag is given to each pilgrim. Pilgrims having smartphone with GPS can use location based services by installing an app. These services include locating family members or friends, requesting urgent help, a map of important locations. To be able to transmit the current position, the app must be running in the phone. RFID readers are also installed in different regions to scan the tags.
The control centre provides features like visualizing the location of all the pilgrims on a map, searching for pilgrim based on several criteria like region, age, etc. Sending notification to the mobile device, maintaining the database of places like hospitals, location history and personal information about pilgrims. Smartphones use web services and the RFID readers use a middleware software to interface with the control centre. The overall architecture is depicted in Figure 3.1.

16. 12
Figure 3.1: Architecture of a system using RFID and Smartphone GPS (Courtesy of [6])
This system faced problems with RFID tags and readers. The read range of the RFID reader was low, it was also affected by the environmental factors. The signal between the RFID tag and the reader was getting blocked by the tag holder’s own body, also the read range was affected by the angle with reader. So they decided to not have a wristband RFID tag.
3.2 Smartphone GPS
A system called “HajjLocator” [1], has been designed to track pilgrims during Hajj. The system uses the GPS built into smartphone. The data transfer is done preferably using Wi-Fi or 3G, but the system can also use SMS to send the data in emergency situations. A pilgrim is required to install an app on his smartphone, which provides several features and sends his location to the server.
3.2.1 Updating Location
The frequency of location update must be decided carefully as it affects the power consumption and the network bandwidth. We can update the location after a certain time period or after the user has moved a certain distance. Choosing time based method to update the location guarantees that we will have the updated location after a particular duration. But if a user is stationary or he is moving slowly, then the frequent location updates will be redundant and will drain the battery and increase the network

17. 13
traffic. If we choose location based method, then we have to continuously keep track of the current position and the last sent position. If the distance between current position and the last sent location is greater than a threshold value, send the current position and store it as last sent location. If a user moves very fast then the distance based method will update more frequently.
The server uses PHP and MySQL. Authorities can login into the web server using a browser to get the position of pilgrims which is displayed in either a tabular form or in Google maps.
Figure 3.2: HajjLocator Architecture (Courtesy of [1])
They conclude that the distance-based update is recommended for tracking pilgrims [1]. Also an option was provided to the user to send his location manually. They also reported that the accuracy in open or semi-opened places was fine, but sometimes the server did not receive any data, why may due to unavailability of a location fix. In some cases, the location obtained from GPS was very far from the user’s actual location. Using the previous location records of a user, these erroneous locations can be adjusted and corrected.
3.3 Wireless Sensor Network (WSN)
In this system each pilgrim is given a sensor unit which contains a GPS for getting the current location of the pilgrim, a microcontroller which executes the program to send the location to fixed sensor units, a battery to power this matchbox sized sensor unit. It also includes ZigBee radio which is used to transfer the data to a network of fixed master units deployed in the region of interest. The tracked mobile units can be much more than fixed sensing nodes. The designers of this system have developed

18. 14
an RFID based system in the past, which provided only the identification of pilgrims. This system is designed to track pilgrims as well as identify them.
The wireless sensor nodes transmit their unique identification number, its current position as obtained from the GPS and the time periodically or on request from tracking station. The data transfer takes place opportunistically using ad-hoc network. The sensor units transmit the data using flooding protocol, so the same data is sent to all the nearby fixed sensors. This data is stored for some time and then multi-hop routing is used to transfer this data. The system designed can tolerate failure of a few fixed sensor nodes. The positions of pilgrims is mapped onto Google map similar to other systems.
For querying the location of a pilgrim, the server sends query by using optimal route utilizing the last known location. The system also supports routing multiple queries in parallel. Battery powered wireless sensor units should have power efficient hardware and software. The data transfer should also be minimum. The increased frequency of updating the location affects power consumption and bandwidth, but it reduces the time taken to find a pilgrim.
3.4 Bluetooth
A number of class 2 Bluetooth devices connected to a host machine are distributed in the area of interest. All of these host machines are networked together. This setup can be simply adding a Bluetooth dongle to a laptop or desktop. A central system which is connected to all these hosts maintains a database which contains the information about the ID of device, the owner of device, etc. This central system also maintains a model of the current positions of all the persons. This information is used to ask a host machine (Bluetooth monitor) to scan for the device IDs of the persons who can be there. For example, if a person leaves one checkpoint, he will probably go to next check point, so the next check point can be ordered to scan the ID of that person’s device.
The system requires frequent connection and disconnection, so the connection time should be minimum. The connection time depends on the clock synchronization between the devices. Once a host knows the clock difference between its attached Bluetooth device’s clock and a Bluetooth device’s clock, it can distribute that information to other hosts. To work this properly, the hosts must have their clocks synchronized. It can tolerate an error of up to 10.24 seconds. This level of accuracy for time synchronization can be achieved by using NTP. The time offset between the hosts as well as their Bluetooth devices’ clock offset needs to be synchronized. This technique enables us to get an expected connection time of 0.64 seconds.

19. 15
3.5 Wi-Fi
Smartphone users are increasing in India. Almost every smartphone comes with at least Wi-Fi. A system was designed for tracking the smartphones on a busy street for traffic flow or congestion monitoring using Wi-Fi. To track smartphones in an area, Wi-Fi monitors with a single radio device are installed. These are standard access points with custom firmware. The power usage of a Wi-Fi monitor was around 7W. The monitors collect the MAC address and the signal strength received. The system can store the data for retrieving it later or it can upload the real time data.
The system takes care of stationary devices also. If a device is present continuously during scan, it is put in a blacklist. If a device is not being observed for a long time, then it is removed from the blacklist. This system used all three techniques to increase the number of frames transmitted as described in section 2.3.2. The active tracking results in 3 to 5 times more packet transmissions compared to passive tracking. The RTS technique has less effect than the others. In their experiment the mean error was under 70 meters and the monitors were 400 meters apart [13].

20. 16
4. SUMMARY
Passive RFID based systems are best suitable for identifying, while active RFID based systems can track pilgrims as well as identify them. Active tags also provide a long read range, but they are more costly. If multiple tags are placed together, sometimes not all of them will be detected. RFID tags operating in ultra-high frequency are susceptible to interference from human body. However, recent ultra-wideband RFID tags are supposed to immunity to interference from the human body and provide long read range. Passive tags compatible with NFC can be the best choice for identifying pilgrims with a smartphone.
Using GPS device on a smartphone can provide more accurate and real time data, but this data needs to be transferred to a central server. According to a study 50 percent of Indian smartphone users do not have data connection. So assuming that a smartphone will always have an Internet connection is not good. Users may disable Internet to reduce the data cost or to save battery. Similarly Bluetooth and Wi-Fi can also be turned off to save battery. So smartphone alone cannot be used to track all the pilgrims. Table 4.1 compares various wireless technologies which can be used for tracking pilgrims.
Table 4.1: Comparison of wireless technologies
Technology
Distance
Power consumed
Cost
Remarks RFID (Passive) < 3 m No battery Very Low Best suited for identification
RFID (Active)
~ 100 m
Very Low
Low
Interference from human body NFC ~ 4 cm No battery High Available only in high end smartphones.
Bluetooth
10 m – 30 m
Low
High
High availability in phones Wi-Fi 20 m – 200 m High High Drains battery of smartphone
GPS + Networking
Anywhere on earth surface
High
Very High
Requires an application to be installed in smartphone
The systems designed so far do not utilize all the sensors available in the smartphone. As Bluetooth is cheaper and available in most of the phone, the systems utilizing Bluetooth in addition to RFID, can be cost efficient. A system can use more than one technologies, for example, having Wi-Fi monitors in addition to Bluetooth monitor can provide more data about the position of pilgrims. The researchers

21. 17
point out the lack of interactivity and facilities in the websites of Hajj [5]. GIS should be incorporated into any pilgrim tracking system. NFC enabled smartphones can be used to scan high frequency RFID tags, which is the most convenient way. It can also be used to quickly setup Bluetooth or Wi-Fi to transfer data.

22. 18
5. CONCLUSION
Using wireless technologies along with smartphones can provide a fast and convenient solution for tracking and managing pilgrims. Using Bluetooth and Wi-Fi capability of a smartphone, along with GPS can provide most efficient solution. Passive RFID tags operating in high frequency which are compatible with NFC, can be read by an NFC enabled smartphone. Active RFID tags are costly which increases the total infrastructure cost as the number of pilgrims are in millions. Practical deployments of RFID based solutions have experienced problems due to body mass and short read range.
Using Bluetooth or Wi-Fi monitors in place of RFID readers can provide more accurate trajectories. Utilizing the built-in Bluetooth or Wi-Fi on a smartphone is cost effective as we don’t have to buy any extra equipment for pilgrims. Also Bluetooth and Wi-Fi monitors are very cheap compared to RFID readers. The location accuracy provided by RFID, Bluetooth or Wi-Fi based solution depends on the number of readers or monitors installed in an area. These systems can only tell whether a person is inside the region of a monitor or not.
The GPS enabled smartphones can provide highly accurate and real time location updates. It requires Internet or other wireless networks to send the location data to central server. Also the battery consumption is high, because an application must be running, GPS device is also on and the data must be transferred using networking. Smartphone applications can be used to access several location based services, like the map of important places nearby, viewing positions of group members on map.
The location data obtained from the RFID reader, Bluetooth or Wi-Fi monitors and the GPS can provide several statistics. The trajectory data of pilgrims can be used for planning traffic, inferring peak hours or restricting the number of pilgrims at a site, providing better crowd management.

23. 19
6. REFERENCES
[1]
T. Mantoro, A. D. Jaafar, M. F. M. Aris and M. A. Ayu, “HajjLocator: A Hajj Pilgrimage Tracking Framework in Crowded Ubiquitous Environment,” in International Conference on Multimedia Computing and Systems (ICMCS), 2011.
[2]
M. Mohandes, “Pilgrim tracking and identification using the mobile phone,” in International Symposium on Consumer Electronics (ISCE), 2011.
[3]
M. Mohandes, “Sustainable Radio Frequency Identification Solutions,” in A case study of an RFID-based system for pilgrims identification and tracking, INTECH, 2010, p. 356.
[4]
M. Mohandes, M. A. Haleem, A. Abul-Hussain and K. Balakrishnan, “Pilgrims Tracking Using Wireless Sensor Network,” in International Conference on Advanced Information Networking and Applications (WAINA), Biopolis, 2011.
[5]
A. Touir, A. Al-Owaisheg and H. Mathkour, “The Architecture and the Design of a Pilgrim Tracking System,” in Information and Communication Technologies From Theory to Applications (ICTTA), Damascus, 2008.
[6]
R. O. Mitchell, H. Rashid, F. Dawood and A. AlKhalidi, “Hajj Crowd Management and Navigation System,” in International Conference on Computer Applications in Technology (ICCAT), 2013.
[7]
S. Hay and R. Harle, “Bluetooth Tracking without Discoverability,” in Location and Context Awareness, Tokyo, Japan, Springer Berlin Heidelberg, 2009, pp. 120-137.
[8]
“Waterproof Silicone RFID Wristband,” Shenzhen Chuangxinjia Smart Card Co.,Ltd, [Online]. Available: http://en.asiadcp.com/company/productinfo.php?pid=60658&upcid=10482&id=128214. [Accessed 28 8 2014].
[9]
R. Want, “An Introduction to RFID technology,” Pervasive Computing, IEEE, vol. 5, no. 1, pp. 25-33, 2006.

24. 20
[10]
S. A. Weis, “RFID (Radio Frequency Identification): Principles and Applications,” MIT CSAIL, 2007.
[11]
Microsoft, “Proximity for Windows Phone: Near Field Communication,” [Online]. Available: http://msdn.microsoft.com/en-us/library/windows/apps/jj207060%28v=vs.105%29.aspx. [Accessed 27 8 2014].
[12]
Nokia, Near Field Communication (White Paper), Nokia Corporation, 2007.
[13]
A. Musa and J. Eriksson, “Tracking Unmodified Smartphones Using Wi-Fi Monitors,” in ACM Conference on Embedded Networked Sensor Systems (SenSys), 2012.
[14]
N. Abedi, A. Bhaskar and E. Chung, “Bluetooth and Wi-Fi MAC address based crowd data collection and monitoring : benefits, challenges and enhancement,” in 36th Australasian Transport Research Forum (ATRF), Brisbane, Australia, 2013.
[15]
Bluetooth Special Intereset Group, “Bluetooth Basics,” Bluetooth SIG, Inc., [Online]. Available: http://www.bluetooth.com/Pages/Basics.aspx. [Accessed 22 8 2014].
[16]
P. Jacobowitz, “GPS and GLONASS: "Dual-core" Location For Your Phone,” Qualcomm, 15 12 2011. [Online]. Available: https://www.qualcomm.com/news/onq/2011/12/15/gps-and- glonass-dual-core-location-your-phone. [Accessed 25 8 2014].
[17]
A. Juels, “RFID Security and Privacy: A Research Survey,” Selected Areas in Communications, IEEE Journal, vol. 24, no. 2, pp. 381-394, 2006.