Usage of a Location-Estimation System - Literature review Example

I will send you the final one with reference list in a few hours. Thanks Research Proposal A location-estimation system using mobile phones, cell towers, and computer network Abstract The proposed research wish to offer an innovative solution to crowd control using mobile technology and computer network. The idea is to find the exact location each member of the crowd or groups and monitor their movement. Since extreme density causes people to push each other and create a force chain leading to crowd turbulence, awareness of the density and movement of the crowd can help organizers control and limit the number of people in a particular area. This proposal specified some details of the upcoming research and explain the rationale of having a location-estimation system using standard GSM cell phones owned by crowd members, four cell towers strategically located around the controlled area, and a local server that would received and process transmitted data. Table of Contents Contents Contents 3 1. Introduction 1.1 General Introduction The Hajj, a yearly event that last for about 10 days, is a ritual of pilgrimage to Mecca where pilgrim joins hundreds of thousands of other pilgrims to perform as series of rituals. Typically, each participant walks counter-clockwise around the Kaaba for about seven times, runs backward and forward between the hills of Al-Safa and Al-Marwah and drink from the Zamzam Well. The pilgrims then stand in vigil in the plains of Mount Arafat and throw stones to simulate Stoning of Devil. Finally, the pilgrims will shave their heads, sacrifice an animal, and observe the four-day Eid-ul-Adha festival. An estimated two million people participated in this yearly event in 2007 thus crowd-control techniques according to Akhter (2009) has been the subject of criticism due to a number of incidents such as crushing as a result of structural collapsed and stampede (p.216). The World Health Organization in January 2010 cited the medical problems associated with Hajj’s large crowd where it finds cardiovascular disease as the most common cause of death. Coupled with heatstroke and severe dehydration, spread of infectious disease due to overcrowding is also a big concern of health authorities (WHO 2010, p.173). The result of one study conducted by Sardar & Badawi (1978) about crowd control during the Hajj pilgrimage reveals the presence of random and irregular crowd flow that requires dimensional analysis and statistical calculations. On suggestion is to use heat flow and random walk simulation to monitor the behaviour of individuals in a crowd as this could reveal the variation of crowd parameters such as velocity (p.113). However, this only part of the solution since the ultimate goal is crowd control. Rahim et al (2011) suggested the use of virtual reality with flexible user control connected to a wireless system for Hajj participants but such system has no ability to control the crowd since their movement still depends on their individual preference (p.68). RFID wristband tag was also proposed by Mohandes et al. (2008) to enable security personnel estimate the number of people in a given area where there is accident. However, the system is generally to control entry of pilgrims in high-risk area through RFID readers installed at key entry and exit points rather than real-time segregation of heavy crowds to smaller manageable groups (p.4-5). More promising, Alnuaim & Almasre developed Location Based Services and Augmented Realty for Hajj pilgrims using mobile phones – Mobile Pilgrim Assistance or MPA. However, although it enhances user interaction and determine user location indoor and outdoor, movement of pilgrims is again dependent on individual location awareness and interpretation of suggestions being shown by MPA (p.1385). Similarly, Yamin & Ades (2009)’s recommendation to combine RFID and Wireless Technology is yet to resolve performance issues when tracking individuals in a very dense crowd such as RFID tracking accuracy and sensor networks sensing and reading in irregular events like Hajj (p.3). Large-scale localization from wireless signal seems more appropriate and as Letchner et al. (2005) demonstrate, such system can effective estimate locations both indoor and outdoor which resolve one primary problem with tracking large crowds. However, the system uses Wi-Fi as location sensor instead of strong GSM cell phone signal, it suffers considerably when tested in remote areas (p.20). This proposed system contain in this research proposal is some form of large-localization using mobile devices with four tracking towers around the pilgrimage site. Information from mobile devices that will be tracked by these towers will be process through a local server enabling real-time crowd control. The plan is to use the power of decision-making and analyzed individual and groups decisions. Idea management for crowd control, which is based on the theory of “wisdom of crowds” developed by James Surowiecki in 2005, can greatly enhance tracking since it can determine the most possible course of action a particular pilgrim or group of pilgrims will take. For this reason, it is important to investigate the viability of the proposed network that will handle this crowd wisdom based tracking system and its ability to actual break and control a large crowd in actual pilgrimage. 1.2 Aims and Objectives The aim of this research is to find a viable solution to manage large crowd during Hajj season in Mecca. The objective is gather enough information to put together a system that can track and control pilgrims movement using mobile devices such as common GSM cell phones and dedicated cell phone towers strategically located around the pilgrimage path. To analyze and evaluate different mobile devices that can be use to facilitate effective and reliable tracking as well as software application, and networking hardware to support the system. 1.3 Problem Statement Hajj is an Islamic annual pilgrimage to the city of Mecca in Saudi Arabia. To date, the estimated pilgrims coming to Mecca is around 4 million thus, the city is increasingly overcrowded despite efforts to control and manage the crowd. In 1975, a gas cylinder exploded in a tent colony killing 200 pilgrims. A riot broke out in 1987 causing the death of more than 400 people while 1, 426 pilgrims died in a stampeded inside Al-Ma’aisim tunnel in 1990. Another stampede killed 270 pilgrims in 1994 while performing the stoning of the devil ritual. There were series of crowd related accidents in early 2000 and in 2006; another stampede killed 346 pilgrims while 289 were injured despite the effort of Saudi government to prevent these tragedies (Gad-el-Hak, 2000, p.56). Beside major expansion in and around the Jamarat Bridge Area, a new system of registrations was recently put in place to control the flow of pilgrims visiting Mecca. However, such system only limits the number of pilgrims by restricting entry of those who already performed the ritual several times rather than real-time crowd control during Hajj (ibid). There were a number of proposals and system developed for crowd management and control such as those mentioned earlier, the decision support system for crowd control of Shubert & Suzic (2007), the SCADA System for Crowd Control by Kamel et al, (2004), the Feedback Technique by Taylor et al. (2007), Mobile Technology for Crowd Management in Cities developed by Alnuaim & Almasre (2010), and the Advanced Mixed Reality Technologies for Surveillance and Risk Prevention Applications by Thalman et al. (2006). Some commercial crowd control management were also introduced and some of them combined mobile digital CCTV and mobile Command Vehicle. Some are with enhanced RFID such as the Intelleflex XC3 Technology, which is based on ISO/IEC 18000-6-2010, and EPC Global CIG2 standards. The Verizon Wireless ‘Mobile Magic’ designed specifically for Disney Parks, Crowd Optic, Colour Mobile VMS, and others. Primarily, the problem this proposal will try to resolve is the occurrence of stampede and other fatal accidents during Hajj. Secondly, since the current system and those that were proposed seems in some part inadequate to realistically control large crowd, the need to investigate, analyze, and develop a system incorporating both crowd wisdom and appropriate mobile and networking technology is required. 1.4 Research Questions The following research questions are formulated using knowledge gathered from different literature concerning crown management and control and analysis of circumstance surrounding Hajj pilgrimage. a) What is the actual problem with the flow of pilgrims performing the ritual in Mecca? b) Why existing crowd control technique is not working? Does it need some innovative approach? c) How can mobile devices help in tracking and controlling the crowd? d) Is breaking heavy crowd possible and viable solution considering the significant diversity of pilgrims? e) Is idea management or ‘wisdom of the crowd’ helpful and applicable in localizing a large crowd? f) Is having an exclusive mobile device network more feasible compared to publicly available systems? g) What are the system components involve? Is it economically feasible? h) Is the proposed towers adequate? How much load can it handle? i) How the system will be tested? What is the sample size required to represent millions of pilgrims? j) Is the system will work in actual situation? How much is the probability of failure? 2. Literature Review One of the main problems in crowd management and control is the pedestrian flow movements and in Jamarat during the annual pilgrimage to Mecca, this problem is very evident (Peacock et al, p.740). Another problem is pilgrim’s behaviour and decision-making ability, which is often the source of crowd panic and subsequent disaster due to confusion and density of the crowd (ibid). One system that was proposed specifically for monitoring and controlling pedestrian at Jamarats’ building and surrounding areas include 700 movable and fixed cameras. Through real-time photography, the system intends to measure and calculate the number of pilgrims in the area, their direction, density, and flow speed. The idea is to limit the entry of pilgrims when a certain area reaches its capacity limit. From this information, organizers are warned to slow the entry of pilgrims or re-direct them to safer site when capacity limits approaches its critical value (ibid). However, it seems difficult to imagine a million people being prevented or re-directed by organizers since this would certainly irritate a good number of pilgrims. More importantly, most pilgrims belong to a certain group and in an event such as re-direction, will typically try to keep their group intact. One noticeable problem with this system is the fact that it will create resistance that is likely to result in contradictory outcome. In their concluding remarks, the designers actually admit the reality that there is no need for physically controlling crowd movement since majority of pilgrims flow are based on awareness movement systems in place such as road signs and safety regulations (ibid, 745). Similarly, the Large-Scale Crowd Density Estimation based on effective region feature extraction method somewhat suffers from lack of real-time crowd behaviour consideration as again, its crown management technique is based on large crowd density estimation and analysis which is only partially helpful in real crowd control situation (Kimmel et al, 2011, p.312). Although clearly incomplete as a crowd control system, but the idea from Voelker (2004), seems beneficial for the proposed system. This author takes into account the rapid individual mobility across network cells and creating of ‘mobile crowds’ which may be helpful in establishing a mobile phone based crowd control system. This ‘mobile crowds’ are people travelling in clusters or groups with mobile phones or devices and accessing cellular networks at the same time. The result is overloading since base stations will be overwhelmed by control traffic. The useful idea here is the fact that a large crowd tracking system that is dependent on mobile devices should have dedicated base stations exclusively for crowd control use (p.64). The four towers mentioned earlier in this proposal are actually dedicated base stations exclusively for the Mecca area serving two purposes. One is to receive mobile phones signals and transmit information to local servers. Considering the human behaviour side of crowd control, let us remember that a crowd is a set of individuals defined by their common social identity (Vadackumchery & Kattakayam 1995, p.165). For this reason, crowd behaviours can be defined as gathering of people’s sentiments and ideas (Le Bon 2006, p.15) with a mind of its own (Plummer 2010, p.34), aggressive (Rouse & Boff 2005, p.492), and sometimes irrational depending on circumstances (Manstead & Hewstone 1996, p.156). In essence, a crowd typically has collective behaviour, which is characterized by similar decisions. From this perspective, it is safe to assume that are indeed strong connection between crowd movement and crowd wisdom. Since the idea is to track a large crowd with mobile devices and break them into small and manageable groups, idea management can play a very important role in predicting and tracking crowd movement. For instance, the commercially available Mobile VMS is taking advantage of crowd wisdom or group decisions by using highly visible and strategically installed signs (Event Industry News 2011, p.1). The idea for this proposal is somewhat the same and the only difference is the method by which information is shown or delivered to the crowd. For instance, consider a GSM cell phone receiving a message “All pilgrims from Pakistan head north” or a PDA with directional map showing pilgrim’s current location and next destination. This approach not only have the advantage of individualized control but available indoor or outdoor which Mobile VMS cannot do. Another advantage is the fact that messages can be delivered to specific groups in a large crowd thereby breaking or reducing crowd density in a particular area. Moreover, provided there is a complete system in place, organizers do not have to physically manoeuvre the flow of human traffic, which may eventually result to dispersal of some group members and panic. This is because the system will inform all members of specific group which direction they should go. The idea of tracking locations of mobile devices is not new. In fact, Letchner et al (2005) employ these devices in large-scale localization project. The same technology was used and tested by Alnuaim & Almasre (2010) in Mina using HTC Touch Viva PDAs with Windows platform and labelled Mobile Pilgrim Assistant or MPA. Although different in approach both projects aimed to provide individual pilgrims with information and guidance using mobile devices such as PDAs and common GSM cell phones. However, these projects seem to ignore the reality of crowd wisdom and the possibility that individuals are sharing ideas resulting to group reactions. They seem so consumed by the technical aspect of crowd control that they merely facilitate tracking and location services. For instance, although the main advantages of these mobile devices are their ability to cover a significantly wide area such a Mecca and transmit from both indoor and outdoor locations, they will be worthless tools if the crowd control managers cannot anticipate and control pilgrims’ movement. Individual and group relationship in a crowd can be found in most crowd model suggesting its importance in determining potential behaviour and movement. For instance, the fractal pattern crowd modelling developed by Widyarto & Latiff (2008) is taking advantage of this phenomenon by using observing interactions of living systems and its ability to create patterns similar fractal patterns found in nature as shown below. Figure 1- Fractal patterns created through interaction The relationship between crowd wisdom and the above fractal is the fact individuals in a crowd interact and create a group with similar motivations and moving in the same direction. The decision-support for crowd control using genetic algorithms with simulation to learn control strategies developed by Shubert & Suzic (2007) also provide interesting facts about human behaviour as affected by hostilities and potential destruction. Although it’s not exactly a representation of a large crowd, it demonstrated the influence of evolving human behaviour and decision-making in its position, direction, and movement as shown in this stochastic agent-based simulation diagram. In the above diagram, one would notice the “combination of actions” as consequence of behaviour and effects. This simulation suggests that individual decisions and motivation play an important role in developing a common approach to problem leading to collective action. Similarly, Donikan (2000) used human behaviour; decision, action, communication, perception, and the environment where individual interact to predict potential crowd movement. These crowd simulation models clearly emphasize the importance of understanding human behaviour particularly those that greatly influence their motivation, decision, and movement. However, since Hajj involves over four million people with heterogeneous groups characterized by internal structure and cultural differences an enhanced approach may be necessary (Malyshkin 2011, p.126). According to Fisher (2009), the tragic event in the City of Mina on January 12, 2006 that killed 346 pilgrims and injured 286 had no apparent cause but features of crowd behaviour in a densely populated environment. The image caught by surveillance cameras provide scientist with data to evaluate the movements of individual pedestrians and small groups at the time of the accident. Surprisingly, the investigation reveals some unexpected features of crowd behaviour that include: Crowd as a whole move continuously regardless of density of people. Pedestrians stop occasionally only when the crowd density is extremely high causing a wave-like stream of people. This stop and go waves was not causing severe problems since pilgrims still feel they are in control of their movement. The real problem starts when the density of the crowd reaches about half square feet per person, as people will be moved involuntary from their current space. As a result, people who were displaced and want to regain their space push others away until it develops a force chain extending to large distances. This force chain then create variations in peoples movement breaking the large crowd into clusters of people moving as a block. Consequently, the intensified force makes weaker pilgrims fall and crushed to death. The best thing to do is to avoid development of force chains. Clearly, the problem associated with Hajj annual pilgrimage is the occurrence of extreme density resulting to force chains, and pilgrims clustered into knots resembling eddies fluid flow (the swirling of a fluid and the reverse current created when the fluid flows past an obstacle) swirling in all directions as shown below. Figure 3 - Eddies fluid flow over and around objects (Source: www.sciencephoto.com) Apparently, the force was everywhere and pilgrims were powerless against the ‘crowd turbulence’, which cannot be predicted by simple crowd models. Dirk Helbing, the lead scientist concluded that it was triggered by individuals changing behaviour in response to crush rather than being pushed along by the crowd. Similarly, crowd density is not the factor responsible for this hazardous turbulence but variations in individual’s movement speed. Both factors are responsible for subsequent crowd pressure (Ball 2009, p.163). For this perspective, it is safe to assume that crowd behaviour is an important factor in determining flow directions and variations in crowd movement speed while controlling density in a certain area prevent the occurrence force chains. Therefore, a crowd control model intended for Hajj pilgrimage or other large-scale gathering for that matter should take into account the potential of controlling crowd density and flow direction, prevent occurrence of force chains, and subsequent dangerous crowd turbulence. 3. Methodology Large-scale localization using mobile devices will take into consideration the power of common GSM cell phones for economic, availability, and signal strength. It will also need the support of at least four standard base stations complete with transceivers, antennas, and encrypting and decrypting data. The local server will also need a server hardware and application that can support GSM network data communication and processing. It will need a fast database processing application. For this reason, the research will make a comprehensive inquiry about GSM mobile phones technology and supporting GSM network infrastructure. It will inquire and study different server hardware and software capable of processing information from GSM signals and store them in the selected type of database. It will also study the Mina and other critical area to establish the appropriate location of GSM cell towers and obstacles that may hinder GSM signal. In order to realize the aims and objectives, the research intends to conduct a literature review of the technical aspects of GSM mobile phone technology including GSM tower components. The purpose is to get familiar with signal and data transmission and equipments that can be use to enable the proposed system to work. These include studying the air interface where the wireless transmission from the cell phone to a base station occurs. The interface between the cell phone and the network base station controller and in particular the visitor location registry or VLR. The potential home network that may be use for the test and the security involve in using such technology. In addition, the study will also study the Subscriber Identity Module, which would identify the mobile. Since the success of the proposed system depends on the number of controlled groups, the research intends to investigate the maximum number of cell phone units that may be accommodated into the network. Similarly, it will also study the variations in indoor and outdoor signal strength and determine the amount of adjustment that may be required by the system. In terms of hardware and software requirements, the research will also do a comprehensive review of available and most applicable hardware and software in the market. The review is in line with the local server requirements and organizer module that will facilitate tracking of mobile phone locations. Another subject of study is the hardware for the server, which may require special processing capabilities as tracking million of cell phones, identification, and segregation of groups, and real-time processing of locations may demand more speed. If not available, the software may be developed for such purpose thus demand further study particularly technical and programming work. Information gathering will be done through informatics database and Internet search. Appropriate and relevant works will be identified through reading and analysis. Those that contain relevant and helpful information will be compiled according to their use in the analysis report. Journal and books available in local and internet libraries such as Google books and others may be use to support arguments and justify analysis. Testing the system requires a representative or test environment almost similar to Mina’s environment, crowd density, and flow. However, imitating such large crowd is nearly impossible thus determination of representative sample is necessary. According to Good & Hardin (2011), determining the optimal sample size require understanding of the smallest effect of experimental significance, the required power and significance level, distribution of the observables, statistic test, and anticipated losses due to non-responding participants (p.32). Consequently, the research should take some time studying these factors and develop an effective strategy that could determine the most accurate sample size that will represent at least a million people in one particular area in Mina. 4. Rational of the research and expected results As the aims and objectives of this proposal suggest, the research generally intend to offer an innovative solution to Hajj pilgrimage’s crowd control problem using a location-based system. The system unlike other experimental model employing PDA, camera, and other commercial available add-on gadgets which may be too expensive for the pilgrims to acquire, will use widely available and commonly used GSM cell phones. Since most pilgrims, regardless of culture, nationality, and economic likely own a regular cell phone, the potential of the proposed system is high. As shown in Fig. 4, the idea is to measure the signal strength of each pilgrim’s cell phone to determine his or her location in a particular area. The purpose of the four base stations is to receive signal from these cell phones and weakness and strength of these signals will indicate the pilgrim’s proximity on a particular base station. Fig. 4 -The Location-estimation system layout The above location-estimation system layout demonstrates the relationship between groups of pilgrims and strategically placed four cell towers. The controlled area, which is represented by the largest oval shape, contained pre-defined or pre-determined location such as buildings as shown by blue rectangle inside the largest oval. Each individual in a group is holding a cell phone and their proximity in any of the four cell towers will determine their location in the area. In other words, those with almost similar signal strength with reference to say “C4” are assumed to be members of a particular group located near C4 as shown above. Note that these cell towers are in strategic locations and exclusive for the area controlled and signals coming in beyond the maximum coverage are ignored. For instance, if the area controlled is 1000 square metres then the maximum signal strength should be within this perimeter and any signal beyond that is considered out of coverage area. With reference to “C3” for instance, signal strength should not exceed the strongest signal received by “C1”, “C2”, and “C4” since this would mean that the person holding the cell phone is outside the controlled area. The purpose of the local server on the other hand is to receive and process data being transmitted by the four cell towers. These data include the location of each cell phone in the controlled area, distance from the reporting tower, and the cell tower code that transmit the data. Organizers monitoring the event can easily determine which group is in what area by sorting the database using signal strength and cell tower code as keys. For instance, same or almost similar level of signals transmitted by C1 means they are near C1 or the specified pre-determined points in the controlled area. Note that the controlled area is geologically mapped with North, South, East, and West with the four cell towers as reference points. In order to break the crowd into manageable groups, organizers can send messages to cell phones with similar signal strength and direct them to a non-crowded area. 5. Time Table Time Table of the Proposed Research No. Task Week1 Week 2 Week 3 Week4 Week 5 Week6 1 Research proposal submission/approval 2 Research preparation 3 Literature review 4 Experimentation 5 Analysis of finding 6 Finalization of research result 7 Submission of draft 8 Final approval 6. Reference List Read More