Principle and Application Progress in Location-Based Services

Compared to vehicle trajectories that are solely generated from outdoor environments, most pedestrian GPS trajectories are recorded in mixed indoor and outdoor environments. Due to the problems of poor indoor accuracy and sparseness of signal points, processing of indoor GPS trajectories is significantly different from that of outdoor GPS data. Existing research often assumes that GPS signal is completely missing in indoor environments. However, with the sensitive GPS receivers and some big windows, satellite signals can also be picked up in indoor environments. To address the above problem, this chapter presents a signal-loss-based method to segment and analyze mixed indoor/outdoor pedestrian GPS trajectories. Firstly, by considering the signal-loss periods in indoor environments, a clustering method is proposed to segment indoor/outdoor sub-trajectories from each trajectory. Based on that, the approach for understanding trajectory patterns is developed, which uses features such as speed, distance and time to recognize “passing” pattern and “indoor activity” pattern in indoor environments, as well as “move-stop” pattern, “more-move” pattern and “more-stop” pattern in outdoor environments. Finally, we evaluate the proposed method with some real trajectories to study its feasibility in segmenting and analyzing mixed indoor/outdoor pedestrian GPS trajectories.

Floor identification is an important aspect of indoor positioning while the resolution of altimetry is not very well, especially in WLAN Location Fingerprinting System. This chapter presents a differential barometric altimetry (DBA) method to identify floor in consideration of features of WLAN location fingerprinting system and the air pressure sensor in a smart mobile phone. The method is that it gets altitude for identification from filtering and calculating the air pressure data which is uploaded real time by both base station and mobile station and the base also support temperature data. The result of experiment shows the resolution of sensor is fairly high and filtered data is steady, the altimetry resolution is about 0.2 m, precision is less than 0.5 m, accuracy is about 1.0 m. All about the experiment indicate the method is fit for floor identification in indoor positioning.

The accuracy of traditional triangulation localization algorithm based on RSSI couldn’t satisfy the indoor navigation applications. This chapter shows a new method added in linear fitting and least square adjustment to solve this problem. In order to achieve better positioning results, we compute the parameters for LDPL model by the linear fitting method and import least square adjustment to the traditional triangulation centroid algorithm. At last, some experiments were carried out in Tongji University using TP-LINK Wi-Fi routers and an Android cellphone. The results show that the proposed method does improve the accuracy about 30 % than the old one, and the Wi-Fi positioning system could be used both in laptop and cellphone applications.

Accurate and trustworthy localization information is crucial to the functionality of a variety of LBS services and applications. However, the infrastructure used for localization, especially indoor localization, is usually vulnerable to signal strength attacks. When under such attacks the localization accuracy usually suffers a great deal of debasement. In this chapter, we focus on achieving robust wireless indoor localization when signal strength attack present on access points (APs). We first investigate the effects of signal strength attack on indoor localization. Then we designed two attack-resistant algorithms to assure the efficiency and validity of the localization information under signal strength attacks. The algorithms can be used to filter out attacker APs, and can be integrated into existing localization algorithms. We testified the algorithms on ICT’s localization engine with an IEEE 802.11(Wi-Fi) environment. Our experimental results demonstrate that our proposed approach can achieve comparable localization performance when AP under signal strength attacks as compared to normal situations without attack.

In recent years indoor localization technology has been regarded as a promising technology. To improve localization accuracy, Inertial Measurement Units (IMUs) embedded in smartphones have been utilized to find landmarks such as corridor, elevator and stairs. This chapter proposes an activity recognition method to identify the landmarks mentioned before. The activity recognition method first determines whether it’s elevator pattern. And then it uses C4.5 algorithm to build a decision tree model to classify walking and taking the stairs patterns. This chapter also discusses the impact of different AR orders and different sample rates to the classifier performance. At last it introduces a real-time activity recognition system based on previous research. The system can recognize activities in about 2 s. In addition, activity recognition and dead reckoning can be used for assisting localization. Compared with WiFi localization technology, this method can evidently save energy at a cost of little localization error.

The ASSISTANT project contributes to maintaining the mobility of older people in Europe, in order to safeguard their social and economic participation in an increasingly ageing society. It does this by helping them to travel safely and independently by public transport. This 3 year project develops an application for the home PC and smartphone that designed to help older travelers to plan their public transport journeys and then receive guidance during their journey. This guidance will help them to find the vehicle they need, warn them when to get off, when and where to change to another route, and will provide assistance if something goes wrong. There are several stages in the guidance where uncertainties play a major role and have an effect on the quality of the trip. The major uncertainty is with the location services when GPS reception in poor or impossible due to urban canyons or the user being under ground or in a tunnel. In addition, when waiting at a stop where for instance several buses might arrive at the same time, it could be difficult to identify the correct bus to board. This paper explains the overall design of the ASSISTANT project and addresses some of the issues related to positional uncertainties.

Radio frequency (RF) signal propagation suffers from time-varying fading effects, and thus radio map-based localization systems are hard to hold the expected accuracy. Base stations (BS)-based architectures show us the probable solutions to overcome the negative impacts by producing adaptive radio maps. In this chapter, the adaptive approach that is presented in our previous work is adopted. To further mitigate the impacts of dynamic environments, we propose a hybrid location estimation method that fuses WLAN signals and inertial data through the sequential importance resampling (SIR) Particle Filter (PF) algorithm. Experimental results suggest that the hybrid method can provide more accurate location tracking, compared to previous algorithms, such as K weighted nearest neighbors (KWNN), initial radio map-based PF, adaptive radio map-based PF, pedestrian dead reckoning (PDR). And it nearly costs equivalent computational time, compared to those radio map-based PF algorithms.

As for the ultra high performance to determine unique earth gravity field model and its geoid, systematic errors and existing outliers need to be removed from the satellite gravimetry observation before scientific product process. In this work, we introduced an improved pre-processing algorithm for satellite gravimetry data. Firstly, scale-factors of observations are calibrated based on certain regional terrestrial-gravity data. Then on the basis of wavelet theory, an outlier-detection algorithm for satellite gravity gradiometry by applying a wavelet de-noising method to some simulation data with white noise and outliers is proposed. The computation result shows that this novel algorithm has a 97 % success rate in outlier identification and that it can be efficiently used for pre-processing real Satellite Gravity Gradiometry data.

Multipath effect is a key factor to perform the precise positioning and navigation with GPS technique. Study on the pseudorange multipath effect at GPS station is conducted in the chapter. Pseudorange multipath effect at the station is computed using the linear combinations of the pseudorange and carrier phase observations. The amount of pseudorange multipath effect at the station is evaluated. Multi-day’s multipath effects are matched based on the repeat time of the GPS constellation. Then the correlation coefficients of matched multi-day’s multipath effects are calculated and analyzed. The results show that there is no high correlation between the adjacent days’ pseudorange multipath and the single-epoch multipath effect is problematic. The reasons for this contradiction include the data matching problem, the effect of the noise, the solving method of the constant part and the effect of tracking error, among which the solving method of the constant part is the most important factor. Then pseudorange observations are corrected with the pseudorange multipath effects. Estimating the kinematic coordinates of the site using the point positioning technique, the positioning from the corrected GPS data is more accurate than that from the raw data.

Geocoded addresses usually code location by a single point that has no attached semantics other than representing the address. The lack of semantics limits the accuracy of any geometric analysis, for example in route planning where the geocode seldom represents exactly the user’s destination. The chapter suggests a semantically and spatially richer address data model, and studies whether this model will significantly improve the accuracy of the geometric analysis in typical location-based services’ tasks. We apply two experiments, one on time gain in navigation, and the other on ranking for k-nearest neighbor queries. Outcomes from both experiments support the argument that location-based services are considerably improved when using the new proposed model, and advantages cannot be simply neglected.

Location-based services (LBS) are becoming an essential part of a person’s personal and social life. LBS service pattern is changing from a location information service to intelligent and personalized user experience build up. Aiming at meeting users requirements and improving the performance of LBS system, this chapter proposes a Location-based Service Cloud Computing System—iWISE. In this system, we emphasize the abilities of location content aggregation and social awareness. We describes our works from following aspects: (1) the architecture of iWISE; (2) the key technologies we implemented in iWISE; (3) a self-adapted campus news recommendation application we developed based on iWISE, and the evaluation criterions for location-based cloud.

The author developed a low-cost data collecting system consists of a laser cross-section scanner, a GPS receiver and an IMU. The chapter demonstrated the functions and characters of all the sensors concerned together with the solution of time synchronization and major procedures of data processing. The system had been used to collect field data and the final experiment results showed that the system reached relatively good accuracy and enjoyed favorable application prospect.

This chapter describes a web-based model of geocollaboration based on geolocalized tweets. From the Twitter stream, it allows selecting those geotagged messages, filtering them by content, and placing them on a map. This map can be zoomed at street level in order to read the displayed messages, or zoomed out to visualize their geographic distribution and frequency. This approach also allows posting messages on the map with an associated geolocalization, other than the user’s current location, corresponding to a specific location where an event is reported or information is requested. Furthermore, all displayed messages can be replied on the map, thus creating a geolocalized conversation thread to support collaboration. Accordingly, our model addresses four main problems: collecting geotagged tweets, geovisualizing these messages, posting and geolocalizing new messages, and supporting geolocalized conversation threading. Experiments showing the promise of the approach to deal with web-based communication issues in catastrophe situations, such as an earthquake or flood, are also described.

As adoption of put information rises, more apps abuse their power to proactively contact their customers. While there are some which are getting this right, only sending notifications when there is something worth getting attention for, others send the same message to its entire user base. Put information is an important service form in the Location Based Service. How to send information to user intelligently in LBS is an important issue. In the chapter, we put forward a way to put information by using the user profile, information types and time, location, environment, and other context information to filter the push content, matches the user demand and potential interested information accurately with the help of both ontology and rules. While ontology describes the acknowledged structured knowledge in LBS, the rules describe declarative knowledge. Then the ontology and rule-based reasoning are used to make implicit information explicit. With the explicit knowledge, we filter the potential information by using users’ characteristic data and the context to achieve precise matching for user needs and information of interests. The example for put information in travel navigation show the method is reasonable and feasible.

Proper initial map reading scale is helpful for improving efficiency in map reading, and helpful for making correct spatial decisions to users. Due to the lack of dynamic link between map reading scale and spatial distribution of geographic information, so the initial map reading scale is often not what users want in need mostly. In order to give user a more reasonable map reading initial scale and to improve the efficiency in map reading, a smart initial map scale method is proposed which connects the initial map scale to spatial distribution of road network based on the analysis of users’ map scale operations. Firstly, the method computes distribution index of road network in different positions with Delaunay triangulation. Secondly, the relationship between spatial distribution of road network and map reading scales is established by collecting users’ reading scale data in different locations. Finally, regression model function of road network and map reading scale are obtained based on regression analysis. The feasibility of the method is verified through smart initial map scale test system in this chapter. The results show that the model can reflect the relationship between the spatial distribution of the road network and map reading scale, also is significant for exploring initial scale in electronic maps.

An important aspect of personal information management (PIM) is the support of our prospective memory, that is, the memory of things to do in future. In particular, calendar-tools or todo-lists help us to keep track of plans and intended actions. Their pro-active capabilities to remind users in appropriate contexts remain limited. To achieve context-dependent and dynamic reminders, this work presents a (1) unifying semantic of various types of activities that allows for aggregation; and (2) a prospective memory formalization. Finally, we introduce the theoretical concept of alert-surfaces to enable context dependent reminders.

The current location-based mobile applications for tourists usually use Web maps as base maps with attached objects like POIs (points of interest) to provide tourists with relevant information, which relies on positioning functions of the current handsets. However, the diversity of maps and geo-information representation methods are insufficient, and are regardless of the differences in cultures as well as target users. The conventional paper-based guidebooks and magazines are still popular because they are good at dealing with subdivided topics, content arrangement, illustrations and stories to provide tentative travel plans with attractiveness and readability. In this chapter, the authors propose a framework to create geo-enabled pages to combine the advantages of positioning-enabled devices and well-designed guidebooks with considering of better user experience in the real world. By analyzing the components of the pages of a guidebook, a structured description of both graphic and geographic information of each page component is established. Different georeference methods are discussed, among which the method of positioning using illustrated-maps is focused. Possible location-based events and interactions with users are enumerated. Finally, a preliminary prototype is developed to test the usability of the framework, followed by a discussion of future issues of this research.

Quick Response (QR) Code is characterized by such features as large quantity of coding information, strong error correcting capability, low cost, etc. In this chapter, the scanning and recognition characteristics of QR code is tested on the basis of realization of indoor passive location system with Windows Phone operating system by taking advantage of the storage space information of QR code, and mutual restrictive relation between related variables, such as scanning deflection, scanning distance, size of QR code, etc. are studied with statistical method. In addition, the outdoor GPS location function and ordinary scanning function of QR code are integrated systematically, so that the outdoor GPS active location is combined to the indoor passive location of QR code, and the space information is combined to attribute information.

Traffic accident base map (TABP) plays important roles in rapid traffic accident treatment. In this chapter, the definition and the contents classification of TABP is studied. Four kinds of contents, traffic signs and symbols, road markings, terrain objects and labels, should be included in such maps. In order to obtain such maps using existing data sources, a method which integrates high resolution satellite images with topographic maps is proposed. Four main steps, preprocessing, geo-reference between imagery and topographic map, symbol and marking digitalization, and field surveying, were introduced accordingly. A mapping system relevant to the TABP generation was developed and applied. At last, two applications, rapid accident treatment and active traffic warning are introduced using such maps.

The aim of this chapter is to examine how mobile positioning data can be used for measuring the impacts of short term events and emergency situations on tourism. As case study, we measure the impact of street riots and political confrontation on incoming tourism with the case study of the Bronze Night riots in Estonia, in April 2007. This political unrest was real emergency situation for Estonia and tourism is one of the most important industries for Estonia. We draw out methodological lessons on using such Call Detail Record based datasets as source for tourism statistics and emergency management.

As an important manifestation of mobile GIS, mobile map has gradually become an important tool which assists people with spatial cognition in modern society. But the mobile information device’s display area is smaller and its operation is more inconvenient than the previous digital device. Mobile map expression with the traditional digital map method may cause the mobile map information imbalance and less readability. This chapter proposed an adaptive variable-scale method of mobile map. Topological relation instead of geometry relation effectively reduced map information imbalance which is caused by the uneven distribution of spatial features. The variable-scale model is chosen by the shape measurement model, and the shape distortion of spatial feature caused by variable-scale model is reduced. The proposed model was evaluated and reliable result was achieved. In addition, a calculation method of mobile variable-scale map loading was presented. The variable-scale mobile map loading and traditional mobile map loading were compared by an experiment. As a result, when the details in the core part of variable-scale mobile map are reserved, variable-scale mobile map information loading is better than traditional mobile map. A new method of mobile map expression is feasible and valid to solve the problem mentioned above.

Being unable to get real-time traffic information ahead makes drivers fail to decelerate before accident happens. Mobile phone locator based road black-spot alarming service system was proposed and developed. Based on mobile phone location data, stopping sight distance model, data smoothing and error correction were presented to integrate traffic safety information. By describing how the mobile phone location information including latitude and longitude data was transformed into plane coordinate, vehicle speed information was available and could be fitted into stopping sight distance model. In the alarming service system, the alarming signal would be sent to drivers judging by the threshold and drivers could take avoiding action as soon as possible. This helps to protect drivers’ personal and property safety. They can decelerate safely before reaching a road accident-prone area or potential accident area.

Location-based services use location as contextual data to exclude irrelevant services from users. However almost all positioning technologies can only provide a location with a certain degree of accuracy. It is necessary to have a framework which can handle this inaccuracy and other uncertainties in order to provide a better and more adaptive service. In addition to positioning inaccuracy, location-based services can suffer from other aspects of uncertainty, such as data incompleteness and inconsistency. There is no universal positioning technique which can provide the position of the user seamlessly indoors and outdoors with an acceptable degree of accuracy. Consequently, it is possible to lose the position of the user for a period of time. To avoid this, some systems use more than one positioning technology, each having incomplete datasets; however they still may produce mutually inconsistent data. If an uncertain spatial dataset is stored and analysed in a framework which cannot handle uncertainty, some aspects of the input data may be missed and the outcome may not be fully applicable in real world applications. This chapter aims at developing a rough set-theory-based navigation application which can provide navigational instructions to users by taking spatial uncertainty into account.

Empirical studies have suggested that rainfall affects travel time. This study presents an investigation of the effects of rainfalls with different levels of precipitation intensity on expressway segment travel time with the variation of traffic flow rate. More than 1 year’s GPS data, traffic volume data and corresponding weather information data from Luoshan expressway segment located in Shanghai Pudong New Area were used for this study. First, a direct method is proposed to obtain the expressway segment average travel time aggregated in 5 min with GPS data from floating cars. Then, a modified BPR function is developed to fit the relationships of average travel time and traffic flow rate under rainfalls with different levels of intensity. The parameters of modified BPR function are calibrated under good weather, slight, moderate and heavy rain conditions respectively. The result demonstrates that average travel times generally get longer as well as more instability when the rainfall is heavier under the similar traffic flow rate condition.