ORIGINAL PAPER

Integrating low-cost RTK positioning serviceswith a web-based track log management system

Daisuke Yoshida & Eugenio Realini &Mirko Reguzzoni & Venkatesh Raghavan

Received: 2 October 2011 /Accepted: 14 January 2013 /Published online: 5 February 2013# Società Italiana di Fotogrammetria e Topografia (SIFET) 2013

Abstract Location-based collaborative platforms are provingto be an effective and widely adopted solution for geo-spatialdata collection, update, and sharing. Popular collaborativeprojects like OpenStreetMap, Wikimapia, and other servicesthat collect and publish user-generated geographic contentshave been fostered by the increasing availability of location-aware devices. These instruments include global positioningsystem (GPS)-enabled phones and low-cost GPS receivers,which are employed for quick field surveys at both profes-sional and nonprofessional levels. Nevertheless, the data col-lected with such devices are often inaccurate. To alleviate thisdrawback, an integration of modern web technologies andonline services with an advanced positioning technique isimplemented. A web-based prototype for quality-based dataselection of GPS tracks, managing track logs and point ofinterests is integrated with the goGPS software. This com-bined system applies the principle of real-time kinematic(RTK) positioning to low-cost single-frequency receivers.The workflow consists of acquiring the raw GPS measure-ments from the user’s receiver and from a network of GPSstations, processing data by RTK positioning through the

goGPS Kalman filter algorithm, sending the accurate posi-tioning data to the web-based system, performing furtherquality enhancements, and logging and displaying the data.Tests were performed in open areas and various dense urbanenvironments, comparing the results obtained by standardGPS devices and by goGPS RTK positioning. Results werepromising and suggest that the integration of web technolo-gies with advanced geodetic techniques applied to low-costinstruments can be an effective solution to collect, update, andshare accurate location data on collaborative platforms.

Keywords Point of interest (POI) . Location-based service(LBS) . Global positioning system (GPS) . goGPS . Quality-based data selection of GPS tracks

Introduction

Background

The latest efforts in technological advancement are gather-ing more and more computational capabilities, wireless con-nection tools, and storage capacity on small devices such assmartphones. Nowadays, most of these devices incorporatealso global positioning system (GPS) chipsets, which allowusers to interact spatially with the Web 2.0 by geo-locatingfeatures and events; this merging between geographical anddigital worlds is giving rise to what is commonly known asthe GeoWeb (Leclerc et al. 2001). One of the main reasonsthat drive hardware-producing companies to embed GPSchipsets into mobile devices is to provide users withlocation-based service (LBS) where wireless connectivityis available. Yoshida et al. (2010) developed a web systemfor managing point of interest (POI) and GPS track log datain order to achieve an effective framework for field survey-ing. The track log management function provided an inter-active interface for log management and data enhancement.

D. Yoshida (*)Faculty of Liberal Arts, Tezukayama Gakuin University,2-1823 Imakuma, Osakasayama City,Osaka 589-8585, Japane-mail: yoshida@la.tezuka-gu.ac.jp

E. RealiniResearch Institute for Sustainable Humanosphere,Kyoto University, Gokasho, Uji City,Kyoto 611-0011, Japan

M. ReguzzoniDIIAR, Politecnico di Milano,P.za Leonardo da Vinci 32, Milan 20133, Italy

V. RaghavanGraduate School for Creative Cities, Osaka City University,3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan

Appl Geomat (2013) 5:99–108DOI 10.1007/s12518-013-0098-4

The POI management function facilitates a collaborativeframework for building POIs with multiple users togetherthrough the Internet.

Tagging web contents with geographic information (i.e.,geo-tagging) has become a common feature on most Web2.0 applications. For example, online photo managementand sharing applications like Picasa1 and Flickr2 geo-reference photos either automatically, using coordinatesstored in EXIF data (metadata associated to digital photos),or by letting users locate them on a map. Also, places andhistorical events described in Wikipedia3 can have coordi-nates associated to them; those through GeoHack4 can bepassed to dozens of other location-aware web services, toget more location-based information. Location can then beassociated to many different kinds of information madeavailable on the web (e.g., blog posts, news, events, shopswith the products they sell, etc.), bridging in this way thedigital and real worlds. This kind of geo-tagging usuallydoes not require highly accurate location data; thus, it isoften sufficient to use a standard low-cost GPS device oreven just clicking on a map. Nevertheless, there are someweb applications for collaborative surveys, for exampleOpenStreetMap,5 that would benefit from getting more ac-curate GPS data, possibly with metadata about their accura-cy level, in order to automatize as much as possible the map-updating process and to provide end users with an estima-tion of the level of error for the data they are going to use.

Low-cost GPS positioning

There are various grades of GPS devices, and the cost ofdifferent devices depends on their hardware quality. Thisresearch targets low-cost GPS receivers, which cost somehundred dollars, in order to lower the cost of highly accuratedata, which could be a crucial element to further widenlocation-based services or business.

There are several issues in using consumer-level low-costGPS receivers such as Garmin, TomTom, or Globalsat prod-ucts. These devices provide only the processed solution inoutput, by means of the de facto standard GPX or NMEAformats, but they do not provide users with the ability toaccess GPS raw data (i.e., GPS code and phase observa-tions). Raw data are employed by these devices for theirown positioning process, normally within black-box algo-rithms, but only the resulting position information is givenin output, together with some metadata information such asthe number of satellites, dilution of precision (DOP) values,and so on. This means that users cannot modify or tune

positioning algorithms to improve the accuracy of the resultfor specific applications.

One of the open source software for enhancing GPSpositioning is goGPS6 (Realini 2009) that utilizes GPS rawdata to enhance the quality of GPS positioning by applyingrelative positioning with respect to a GPS master station.This software has the potential to achieve the accuracy atsub-meter level such as 40~80 cm (Pertusini et al. 2010;Yoshida 2011). In this research, goGPS was adopted as apositioning engine with low-cost GPS receivers since thepositioning process can be made available as a standardizedweb service (Yoshida 2011; Realini et al. 2012) to be usedfor various applications and services.

goGPS can significantly mitigate GPS errors (e.g., clocksynchronization errors or atmospheric effects on the signal)and enhance positioning accuracy. However, the softwareneeds to be installed on PCs, and the operations are cum-bersome for users. Therefore, a web-based prototype systemfor track log management system (Yoshida et al. 2010) wasintegrated with goGPS positioning to support goGPS datamanagement and also to provide a collaborative frameworkfor multiple users. Further, a new quality-based data selec-tion method for effectively eliminating inaccurate track logpoints, especially, cased mainly by multipath in dense urbanenvironment. The method used an index based on theKalman filter (KF) error covariance matrix (Kalman 1960;Grewal and Andrews 2001), and some tests were conductedto clarify the effectiveness of the developed index in thisresearch.

This research investigates three aspects of the improve-ment of low-cost GPS positioning:

1. Applying single-frequency (L1) real-time kinematic(RTK) positioning to low-cost devices by goGPS opensource software

2. Integrating a web-based track log management systemwith goGPS positioning service

3. Implementing a suitable index for filtering out poorlypositioned points

System architecture and implementation

goGPS architecture

goGPS is a MATLAB7-based open source software packagethat gets raw GPS data (code pseudorange, carrier phase,signal-to-noise ratio, ephemeris, and timing) in input andprocesses them within a KF specifically developed to applyRTK and address low-cost GPS navigation issues. goGPS1 http://picasaweb.google.com

2 http://www.flickr.com3 http://www.wikipedia.org4 https://wiki.toolserver.org/view/GeoHack5 http://www.openstreetmap.org

6 http://www.gogps-project.org7 http://www.mathworks.co.jp

100 Appl Geomat (2013) 5:99–108

KF can optionally include additional sources of data such asdigital terrain models or line networks in order to improvethe final positioning. goGPS receives data both from a low-cost receiver through a COM port and from a master station,belonging or not to a network of permanent GlobalNavigation Satellite System (GNSS) stations, through theInternet via NTRIP protocol (Fig. 1). At the moment,goGPS is capable of retrieving GPS raw data for the rovingreceiver from u-blox LEA modules (“T” version), FastraxIT03 modules, and SkyTraq S1315F-RAW modules, bydecoding their proprietary binary formats; as for the baseGNSS station data, the RTCM 3.1 format is supported.goGPS can run either in real-time or post-processing mode,synchronizing the rover and master data streams and man-aging temporary outages or permanent data losses. Variousevents such as satellite additions/losses, change of pivotsatellite (highest satellite used for double differences), andcycle slips are managed.

goGPS main targets are single-frequency low-cost devi-ces, but its code and phase modules are designed to workeither in single-frequency or dual-frequency mode.Therefore, it can be used also with dual-frequency receivers,if it is needed. goGPS includes also an alternative version ofits main Kalman filter algorithm, designed to obtain line-constrained positioning (e.g., to navigate on a network ofroads). For a detailed description of goGPS Kalman filter,see Realini (2009).

Despite the latest technology, trends are shifting fromsystem-on-chip (SOC) architectures to a host-based orserver-based one; the majority of people (and some manu-facturers), when dealing with GPS positioning engines, stillthink in a SOC direction. Although goGPS was not explic-itly designed since the beginning with this distinction inmind, it has naturally become a host- or server-based systemfor at least three reasons:

1. It has been first developed as a tool for studying andteaching GPS positioning in a MATLAB environment.

2. Applying RTK requires access to raw GPS observa-tions, which means at least time-of-week, code pseudor-ange, and carrier phase.

3. It has been developed with the concept of free and opensource software (FOSS) and open data policy as itsfoundations. In order to let developers have the greatestfreedom over how to implement their own positioningengines, GPS raw observations must be available, andtypically, they will be processed on their host systems orservers.

In order to make the goGPS positioning open andwidely available in various applications or services,Yoshida (2011) and Realini et al. (2012) implementedthe goGPS positioning process as a web service com-pliant to international standards. The definition of web

processing service (WPS) is provided among the OpenGeospatial Consortium (OGC) Web Services (OWS)specifications (The Open Geospatial Consortium2003). OWS are key elements that can provide effi-cient ways to distribute data and processing results tovarious GPS-enabled clients. The goGPS WPS-compliant positioning service was implemented usingZOO8 (Fenoy et al. 2012), one of the available opensource WPS engines, which provides a standard pro-cessing service through the web. Figure 2 shows a webinterface for goGPS positioning using OpenLayers,9

which can select and upload the required GPS datasetsin input and send a WPS request to ZOO for executinggoGPS. The web client shows the result that isreturned from the server as the white line in Fig. 2.In this example, three RINEX files are used as inputdatasets: an observation file containing raw data for theGPS rover (JENOBA_test_03_rover.obs), a navigationfile containing satellite ephemeris and ionosphereparameters (Vb20.10n), and an observation file contain-ing raw data for the master station (Vb20.10o). Byaccessing the URL from a browser, the RINEX files,which were previously uploaded on the server by theweb interface (Fig. 2), are parsed to extract the GPSraw data; the GPS observations are then processed bygoGPS, and the result is stored in a KML file which isdownloaded through the browser itself after a fewseconds from having issued the URL request. In addi-tion, the displayed tracks can be stored in a databaseby clicking “Store the GPS track on the server” link inthe OpenLayers interface and can also be searched inthe POI management system that is explained in thefollowing section.

Web-based track log management system architecture

Yoshida et al. (2010) developed a web-based track logand POI management system that can import, manage,and display GPS track logs together with POIs such asgeo-referenced photos and videos. The system was in-tegrated with the goGPS positioning engine as describedin the previous section in order to provide higher accu-racy of positioning and interoperable service. The archi-tecture of the system can be divided into threecomponents (Fig. 3). The first is the GPS componentthat collects GPS raw data to be sent to the server forprocessing. The second is the server component thatprovides data archiving and the goGPS positioning ser-vice, which was shown in Fig. 2. The third is the clientcomponent for viewing GPS locations and track logs. In

8 http://www.zoo-project.org9 http://openlayers.org

Appl Geomat (2013) 5:99–108 101

particular, the server component enables searches withintrack logs and tracking services to the client compo-nents through HTTP.

The system connects a Flickr photo album with a GPStrack log by implementing the Flickr web API, and it alsoprovides its function as a web service. The implemented webservices are based on standard HTTP communication, so thatthe user does not need to install any special software in orderto use them. Service requests can be made directly through abrowser, making the web services straightforward and easy touse. These open web APIs help reduce redundant work anddevelopment costs. The URL below is to search GPS track

logs from the database. The user inputs the parameters such asdate, time interval, or DOP to search and filter the GPS trackdata. The request URL returns an XML-like document as aresult that displays the track logs in OpenLayers interface.

http://SERVER_ADDRESS/DP/filtering_simplify.php?address=%25&stime=00%3A00%3A00&etime=23%3A59%3A00&sdate=2009-08-12&hdop=3&gpsmode=3

Fig. 1 Overview of goGPSdouble difference observationsusing a GPS master station

Fig. 2 Positioning resultdisplayed in the goGPSweb interface

102 Appl Geomat (2013) 5:99–108

The system allows users to query and display tracklogs interactively by selecting date, time, IP address ofGPS tracking components, GPS positioning status (i.e.,2D, 3D fix, and no fix mode), number of satellites, andDOP on OpenLayers web interface.

POI management with Flickr service

Recently, many increasingly popular web applicationsfor building POIs, such as Flickr and Google PicasaWeb Album, have been implementing web-mappingfunctions. Also, a popular social media service,Twitter, provides various web APIs with geographicalsearch functions.

Taking pictures using GPS-enabled phones or cam-eras can save location information in an EXIF imagefile, a process commonly known as “geo-tagging.” EXIFis the standard for storing metadata in image files cre-ated by digital cameras, such as date, time, shutterspeed, focal ratio, etc. The POI management systemsynchronizes non-geo-tagged POIs with GPS track logson the basis of the common date and time attributes.Figure 4 shows a POI (as a red marker) and the infor-mation (in this case a photo, the date and time it wastaken, owner’s name, and tag words) in a pop-up box.

By using this system, users are not required to carryspecial devices for field surveys such as professionalGPS and GPS-enabled phones or cameras. The systemalso supports the use of geo-tagged images to extractthe embedded location information.

The results displayed in the OpenLayers interface, theGPS points and POIs, can be exported as KML and GPXformats for further use on other GIS applications. For ex-ample, an exported KML is displayed in Google Earth inFig. 5. Recently, since OGC has adopted the OGC KMLEncoding Standard (The Open Geospatial Consortium2007), many applications have supported KML, among

which are ESRI ArcGIS,10 Flickr, Yahoo! Pipes,11 NASAWorldWind,12 and Microsoft Bing Maps.13 Therefore, KMLis provided by the system here described as a good solutionfor letting different applications interoperate.

Quality-based data selection

The web-based track log management system has on-the-fly functions to filter GPS data by quality-relatedparameters such as number of satellites, DOP, and fixmode. “Fix” is a term commonly used for consumerGPS devices, which indicates the action of computinga position (e.g., “time to first fix” or “TTFF” indicatesthe time the receiver needs to compute a position afterhaving been switched on). This is not to be confusedwith the fixing of ambiguities, which are typically re-ferred to as “fixed” as opposite to “float.” GPS posi-tioning needs at least four satellites to calculate a “3Dfix” position, which includes altitude. The calculationusing three satellites performs “2D fix” positioning.Some receivers even record GPS data under “no fix”status (e.g., the quality of the signal was lower than achosen threshold, so the positioning is labeled as “nofix”); thus, many noisy positions can be displayed onthe web map. The web system provides an interactiveweb interface to check the parameters and results at thesame time in order to remove such noisy data from theplot. Figure 6 shows the effectiveness of horizontalDOP (HDOP)-based data selection, on a dataset sur-veyed at car speed in an urban environment. Figure 6aindicates the result without HDOP-based data selection.It displays a lot of noisy data at the beginning pointbecause the GPS receiver was just switched on, and itwas not getting good geometries of satellites at themoment. In contrast, Fig. 6b shows a HDOP-based dataselection result, in which noisy data were reduced andpoints were more defined.

Kalman filter DOP

Using the HDOP as an index for removing poorlypositioned points has the major drawback of taking intoaccount only the satellite geometry. This informationcan be useful if the sky visibility condition is alwaysthe same, but in urban environments, the actual posi-tioning accuracy is highly dependent on the signal qual-ity, especially when using high-sensitivity GPS devices.This means that the receiver can track a high number of

10 http://www.arcgis.com/home/11 http://pipes.yahoo.com/pipes/12 http://worldwind.arc.nasa.gov/java/13 http://www.bing.com/maps/

Fig. 3 The three components of the track log management system

Appl Geomat (2013) 5:99–108 103

satellites, maybe also with a good geometry, even insituations where the signal is highly degraded: in thiscase, the HDOP would not be a good indicator of theactual positioning quality. To overcome this limitation,goGPS does not compute only usual DOP values, whichdepend only on the satellite geometry, but it producesalso customized DOP indices based on the actual qual-ity of the position estimation. According to the defini-tion given by Spilker (1996), DOP values originate

from the coordinate cofactor matrix Q. This is obtainedfrom the least-squares design matrix A as (Hofmann-Wellenhof et al. 2007)

Q ¼ ATA� ��1 ð1Þ

Thus, for example, the traditional position dilution ofprecision (PDOP) is

Fig. 4 POI and GPS track logdisplayed in OpenLayersinterface

Fig. 5 Displaying exportedKML using Google Earth

104 Appl Geomat (2013) 5:99–108

PDOP ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiqX þ qY þ qZ

p ð2Þ

where qX, qY, and qZ are, respectively, the X, Y, and Zweights along the main diagonal of Q. In order to haveindices that represent how the filter is performing, we ex-ploit the coordinate error covariance matrix C estimated bythe Kalman filter, i.e.,

C ¼ I � GAð ÞK ð3Þ

with I as the identity matrix, G the Kalman filter gainmatrix, and K the coordinate error covariance matrixbased on dynamics only. More specifically, the co-variance matrix K is computed by propagating thecoordinate error of the previous epoch to the currentone through the Kalman filter transition matrix andthen adding the contribution of an a priori-definedmodel error. However, what is really crucial here isthat the gain matrix G depends on the observationerror covariances that in goGPS are modeled not onlyas a function of the satellite elevation but also of thesignal-to-noise ratio (Realini 2009). The PDOP valueobtained from the Kalman filter coordinate error co-variance matrix (from now on KPDOP) is computedas

KPDOP ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffifficX þ cY þ cZ

p ð4Þ

where cX, cY, and cZ are, respectively, the X, Y, and Zvariances along the main diagonal of C.

As for the Kalman-based horizontal and vertical DOPvalues, respectively KHDOP and KVDOP, they are com-puted by following the same logic of traditional HDOP andvertical DOP (VDOP), i.e., propagating the covariance froma global (X, Y, Z) to a local (east, north, up) reference frameby the equation

CENU ¼ RCRT ð5Þ

Fig. 6 Result of HDOP-based data selection in OpenLayers interface a before data selection and b HDOP data selection result

Fig. 7 Complete tracks of experiment (circles) are displayed

Appl Geomat (2013) 5:99–108 105

where R is the rotation matrix from global to local frames,thus obtaining

KHDOP ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffifficE þ cN

p; KVDOP ¼ ffiffiffiffiffiffi

cUp ð6; 7Þ

where cE, cN, and cU are, respectively, the east, north,and up variances along the main diagonal of CENU. Inthis way, we obtain alternative DOP indexes that canbetter describe the positioning quality obtained by theKalman filter. K*DOP values do not depend exclusivelyon satellite geometry, but also on the evolution of thefilter itself, which includes for example the varianceincrement for slipped satellites.

goGPS and the web-based track log managementsystem have been made interoperable by addingNMEA output capability to goGPS and by tuning theparsing algorithm on the web-based system. Since theNMEA 0183 specifications include the possibility todefine customized (usually vendor-specific) sentencesto provide additional information not encompassed by

standard sentences, a customized sentence was definedto output KPDOP, KHDOP, and KVDOP values inNMEA format.

Tests for quality-based data selection

Some tests were performed in order to check theperformance of the KHDOP index compared to thestandard HDOP and to verify the feasibility and per-formance of raw data positioning in low-density andhigh-density urban environments. The first examplepresented here, surveyed by car in a low-density urbanenvironment in Italy, was chosen because part of itwas recorded on the first floor of a two-story parkingstructure, providing a sudden change from good skyvisibility to no sky visibility at all. Since the GPSdevice used was a u-blox AEK-4T, its high sensitivityallowed for signal reception and positioning even withthe highly degraded signal inside the parking structure.

Fig. 8 GPS tracks are selectedby DOP. a HDOP-filtered track,b KHDOP-filtered track

Fig. 9 HDOP versus KHDOPfor the whole surveyingtime span

106 Appl Geomat (2013) 5:99–108

Figure 7 shows the complete track without data selec-tion (the parking structure is in the bottom right-handside of the image), while Fig. 8 shows the comparisonbetween HDOP-filtered and KHDOP-filtered tracks.

The positioning inside the parking structure was verybad due to the lack of sky visibility, but the HDOP-filtered track (Fig. 8a) keeps some of its points evenwhen the threshold is very low. By using such a strictthreshold on HDOP, based on satellite geometry only,part of the outdoor track is deleted although it is muchmore accurate than the indoor track. On the other hand,the track filtered by KHDOP (Fig. 8b) effectivelyremoves only the points surveyed within the parkingstructure, leaving unchanged the rest of the track.

Figure 9 shows the differences in the two DOP values forthe whole survey. The Y-axis indicates the level of theDOPs, and X-axis is for GPS time. In the graph, it is clearthat KHDOP describes well the quality of the GPS signal

under poor conditions (e.g., inside parking, under trees, orsome other obstacles), thus for discriminating between poor-ly positioned points and points with higher accuracy. Othertests have been performed, confirming the better perfor-mance of the KHDOP over the HDOP as an indicator ofthe positioning quality.

The other test was conducted in a high-density urbanenvironment in Japan. Figure 10 shows a path, taken bycar in Osaka and Sakai cities with a TOPCON dual-frequency receiver, which indicates good sky visibility asyellow and bad visibility as red. The result of the test wasshown in Fig. 11, comparing the HDOP and KHDOP valuesin the “good” and “bad” sections in terms of mean, standarddeviation, and RMSE. The plots clearly show that KHDOPindicates a more significant difference than HDOP between“good” and “bad” sections, indicating that it may be moresuitable for determining the positioning quality when usinghigh-sensitivity receivers in an urban environment.

Fig. 10 “Yellow” sectionsindicate good sky visibility, and“red” ones show bad visibilityin Osaka and Sakai cities,detected by a TOPCONdual-frequency receiver

Fig. 11 HDOP versus KHDOPin mean, standard deviation,and RMSE

Appl Geomat (2013) 5:99–108 107

Conclusions

In this article, a web system that utilizes the goGPSpositioning service is described, also showing how otherservices can be connected or merged to it, giving rise tothe so-called “Mashups” or “Chains.” The integration ofweb technologies with GPS positioning enhances inter-operability of the positioning service. In particular, inthis research, goGPS positioning services and a web-based track log management system are made interop-erable in order to increase the usability of goGPS; theaccuracy of the final result is enhanced by introducing anew technique to disregard low-quality position esti-mates, relying on Kalman filter-based DOP values(KPDOP, KHDOP, and KVDOP), which are effectivealternatives to traditional DOP values.

The integrated system does not only facilitate a col-laborative framework for building POIs, but alsoimproves accuracy and reduces costs in current GPSreceivers. The outcomes of this research can contributeto make highly accurate location data closer toconsumer-level users by providing open and standard-ized positioning services. Finally, the obtained resultsshow that RTK positioning based on low-cost receiverscan guarantee good accuracy and reduce costs ondeploying high-quality LBS.

Acknowledgments The authors acknowledge the positioning serv-ices IREALP GPSLombardia (Italy) and JENOBA (Japan) for theirsupport. This research was supported by the JSPS Grant-in-Aid forScientific Research (issue no. 2109737) entitled “Development ofUbiquitous LBS Web-Service using Free and Open Source Software.”The authors would like to express their gratitude to Dr. TerenceHenares, researcher at Osaka Prefecture University, for carefullyreviewing this paper.

References

Fenoy G, Bozon N, Raghavan V (2012) ZOO project: the open WPSplatform. Applied Geomatics. doi:10.1007/s12518-011-0070-0

Grewal MS, Andrews AP (2001) Kalman filtering: theory and practiceusing Matlab. Wiley, New York

Hofmann-Wellenhof B, Lichtenegger H, Wasle E (2007) GNSS GlobalNavigation Satellite Systems: GPS, GLONASS, Galileo, andmore. Springer, Wien, pp 262–266

Kalman RE (1960) A new approach to linear filtering and predictionproblems. Transaction of the ASME-Journal of Basic Engineering82:35–45

Leclerc Y G, Reddy M, Iverson L, Eriksen M (2001) The GeoWeb—anew paradigm for finding data on the web. In Proceedings of theInternational Cartographic Conference

Pertusini L, Realini E, Reguzzoni M (2010) goGPS: accurate roadmapping using low-cost GPS receivers. In Proceedings of GIS-IDEAS 2010, Hanoi, Vietnam 5: 67–72

Realini E (2009) goGPS—free and constrained relative kinematicpositioning with low cost receivers. Dissertation, Politecnico diMilano

Realini E, Yoshida D, Reguzzoni M, Raghavan V (2012) Enhancedsatellite positioning as a web service with goGPS open sourcesoftware. Applied Geomatics 4(2):135–142

Spilker JJ (1996) Satellite constellation and geometric dilution ofprecision. Global positioning system: theory and applications, 1,American Institute of Aeronautics and Astronautics (AIAA). ProgAstronaut Aeronaut 163:177–208

The Open Geospatial Consortium (2003) OpenGIS® web processingservice (WPS) implementation specification version 1.0.0. http://www.opengeospatial.org/standards/wps/. Accessed 7 Aug 2011

The Open Geospatial Consortium (2007) OpenGIS® keyhole markuplanguage (KML) implementation specification version 2.2.0.http://www.opengeospatial.org/standards/kml/. Accessed 7 Aug2011

Yoshida D (2011) Development of processing services for GPS accu-racy enhancement by data filtering and kinematic relative posi-tioning. Dissertation, Graduate School for Creative Cities, OsakaCity University

Yoshida D, Song X, Raghavan V (2010) Development of track log POImanagement system using free and open source software. AppliedGeomatics 2(3):123–135

108 Appl Geomat (2013) 5:99–108