GPS Basics & Status of GPS in Qatar Charles A. Chamberlain, President, Chamberlain Consulting, Ottawa Canada | Abstract Of The Paper & The Profile of The Speaker | Speaker Index | Paper Title Index | Abstract In 1993 it was decided to use GPS as the primary tool for Surveying and Mapping activities in the State of Qatar. Several tasks were undertaken over the next three years to support that decision, including: Development of the 1995 Qatar National Datum QND95 as the underlying coordinate base for positioning applications. Development of the 1995 Qatar Geoid Model (Qatar95). Establishment of a permanent GPS Base Station which operates on a continual basis. Development of an on-line Bulletin Board Service for dissemination of GPS observations. Development of a Real Time Differential GPS System (RTDGPS). Development of Microsoft( Windows compatible software for transforming GPS based WGS84 coordinates to QND95 and computing geoid model parameters. Design, installation, and implementation of special GPS Validation Networks for validation of GPS equipment, user procedures, and calculations. Providing one, two, and five day education and training seminars on GPS. The how and why of each of these activities is described in this paper. Where necessary some basic GPS theory is given to clarify the discussion. The paper concludes with a description of how GPS is used on a day-to-day basis to accomplish the surveying tasks of the Centre for GIS. 1.0 Introduction Geodetic networks provide the underpinning for the Surveying and Mapping activities of a country. They provide a precise and consistent reference system so that all users have a common coordinate base and can readily share data between their applications. Qatar's geodetic network was first established in 1970. It consists of 95 stations. Classical first order terrestrial techniques, distance, direction, and azimuth observations, were used for the survey. Other networks, such as the second order Greater Doha network and third order densification have resulted in approximately 5000 geodetic stations in Qatar. Increasing accuracy demands, easier access, and real or near real time positioning requirements have made Qatar's geodetic network less useful. Large scale mapping, modern vehicle location requirements, and offshore positioning could not be supported by the existing network. In order to support these increased demands, it was decided to upgrade the current network to ensure that it could and at the same time introduce the use of the Global Positioning System (GPS) into the operations of the Centre for GIS. Through the use of GPS all foreseeable uses of the geodetic network and related positioning requirements could be realized. In 1974, the United States Department of Defense (DOD) began implementation of a satellite ranging and timing system known as the NAVSTAR (Navigation System with Time and Ranging) Global Positioning System, or GPS for short. In June of 1993 the final satellite constellation of 21 active and 3 spare satellites was completed. It provides for three dimensional positioning, 24 hours per day worldwide, to an accuracy of 100 metres using the Standard Positioning Service (SPS). With Selective Availability (SA) turned off, it is possible to achieve real time positions to 25 metres. (Selective Availability is a method used by the Department of Defense to degrade the accuracy of the signals in a time of military emergency). Although GPS was designed as a military system civilian users far outnumber military users. Using specialized observation techniques relative positioning accuracy has been reduced to the centimetre level with observations times as short as a few seconds using kinematic techniques and to 1-3 parts per million (ppm) of the line length using static techniques. Relative static techniques, where two or more GPS receivers are stationary at geodetic network points for a period of time, were used to re-survey the Qatar first order geodetic network. Two networks were designed and observed, a "Basic Network" of nineteen stations plus a permanent GPS Base Station and a "Re-observation Network" of seventy nine stations which included the Basic Network. Existing G stations were used for all stations except for the new GPS Base Station which was established on the roof of the Centre for GIS's main office. To relate Qatar's geodetic network to the global reference system, ties were made to an International reference station in Bahrain. Heights computed from GPS surveys are geometric distances above a reference ellipsoid. These "ellipsoidal heights" are of little use in the real world where heights above the "geoid" or mean level are used. Heights above the geoid are correctly called "orthometric heights". A geoid model provides the difference between ellipsoidal and orthometric heights, the "geoid ellipsoid separation". A geoid model, Qatar95 was developed. GPS surveys provide geodetic coordinates on the World Geodetic System 1984 (WGS84) datum. In order to relate the WGS84 coordinates derived from GPS to the Qatar National Datum (QND) coordinates used in Qatar a set of coordinate transformation parameters had to be computed. Because it is not possible to compute an errorless transformation between WGS84 and the existing QND datum, it was decided that a better approach would be to define a new QND datum that has an exact relationship to the WGS84 datum. The new datum is the Qatar National Datum 1995, or QND95. Establishment of the GPS Base Station has two distinct advantages for GPS users. Firstly, it ensures that users of the Base Station data will derive coordinates from GPS surveys that are on the QND95 datum. This advantage relates to one of the fundamental reasons for using control networks. The second advantage is that all users of GPS in Qatar are provided with an extra receiver for surveying. For example, differential positioning requires two receivers, one at a known location and one rover. The known location receiver is provided by the Base Station so the user only needs only one receiver. To compute a differential position, data from both receivers are necessary, therefore, a Bulletin Board Service (BBS) was developed to disseminate the data to users electronically. The BBS is currently being converted to an Internet web site. To ensure the successful implementation of any new technology, training and education are mandatory. Training modules have been developed and presented to both Centre staff and the general public. The remainder of this paper describes the work that was undertaken to implement GPS in the State of Qatar and the results achieved. It then concludes with a description of how GPS is being used to achieve the day to day survey tasks at the Centre for GIS. 2.0 Geodetic Networks The first step undertaken was to re-establish the horizontal geodetic network. Two networks were designed, a "Basic Network" and a "Re-observation Network". Each network had its own objectives and requirements. 2.1 Basic Network The objectives of the Basic network were to: Establish a permanent GPS tracking station (Base Station) to be used for GPS operations in Qatar, Establish the WGS84 coordinates of the Basic network points by tying into the US Defense Mapping Agency station in Bahrain, Determine the datum transformation parameters between the WGS84 datum and the Qatar National Datum (QND), Determine a Qatar geoid model, and Train Centre for GIS staff in GPS technology and techniques. 2.1.1 GPS Base Station A permanent GPS Base Station was established on the roof of the Centre for GIS. This location provided several benefits: It is easily accessible, It provides a secure environment for the GPS equipment, There is ready access to an AC power source, There is convenient access to the Centre's computer network, and It is a suitable location for a differential positioning Base Station and for a real time Base Station. Installation of the Base Station was accomplished in mid May 1993. The point is a five-eighths inch threaded bolt cemented into the parapet wall on the roof of the building. A bolt was chosen so that a GPS antenna could be attached directly to it. A cement block enclosure was built for protection of the station, power was installed, and conduit was installed to protect all cables running into or out of the building. Existing stations were chosen for the remainder of the Basic Network. By choosing existing stations it was not necessary to install any new monuments. Stations were chosen on the following basis: Accessibility, Stability, Availability of reliable height information, and Location in existing network to ensure a uniform distribution of the entire country. 2.1.2 Ties to IGS Station in Bahrain The second major objective of the Basic network was to establish the WGS84 coordinate system in Qatar. The US Defense Mapping Agency (DMA - note DMA is now called National Imagery and Mapping Agency - NIMA) maintains a station in Bahrain which is part of the International GPS Geodynamics Service (IGS) network. Arrangements were made with DMA to continuously collect observations at Bahrain during the establishment of the Basic network. Observations were later received from DMA and processed together with those collected at the Basic network stations. 2.1.3 GPS Network Design GPS network design consists of two steps: (1) determination of the GPS observation technique to be used, and (2) selection of lines to be observed. In order to obtain the highest possible accuracy, the static GPS survey technique was chosen for the observations. With this technique two or more GPS receivers are set up at the network stations and remain stationary for the duration of the observation interval. One of the receivers must be at a known location and the "relative" difference between the receivers is determined. Typical observation periods range from one to two and a half hours depending upon the length of the line and accuracy requirements. Table 1 provides recommended observation intervals for static surveys. Line Length Observation Interval 15 - 50 km 60 - 90 minutes 50 - 100 km 90 - 150 minutes > 100 km > 150 minutes Table 1 Static Observation Intervals Line lengths for the Basic network varied from 5,631 metres to 154,760 metres with an average length of 47,618 metres. It was decided that due to the importance of the network all observation sessions would be 3 hours long. The DMA station in Bahrain provided the known position. The observation epoch was set to 15 seconds. Observations were planned to last from May 18 to May 31 1993. The Base Station was turned on each day from approximately 5:30 to 17:00. When a permanent connection was made to an AC power source on May 24, the Base Station tracked continuously. As part of the network design each station was to have a minimum of 6 hours common observations with the Base Station and 3 hours common with its neighbors. At the completion of the survey on June 6th the following observations had been collected: Station Data Collected Bahrain 336 hours of 30 second data Base Station 405 hours of 15 second data Network Stations 207 hours of 15 second data Table 2 - GPS observations for Basic Network An average of 10.9 hours data was collected at each network station. 2.2 Re-observation of the Geodetic Network It was decided to re-observe the entire existing first order network. Objectives of the Densification network phase were: Determine WGS84 coordinates for the remainder of the stations, Compute the WGS84 to QND datum transformation parameters, and Verify the Qatar95 geoid model. Qatar's first order geodetic network is made up of 95 stations. A reconnaissance was performed to find all of the stations. Only 54 of the existing G stations were recovered or usable. In addition, 5 stations in the second order Greater Doha network were reconnoitered. Three hour static GPS observations were made on these 59 new points (15 second epochs). Adjacent stations were observed together with the Base Station and the Basic Network. Approximately 45 days of observing time were required to make all the observations. A total of 732 hours of observations were collected at the network stations (An average of 9.4 hours per station). The Base Station observed continuously, 24 hours per day. 3.0 Qatar Geoid Model GPS determined coordinates are with respect to a three dimensional Cartesian coordinate system {X Y Z} with its origin at the centre of gravity of the earth. In order to work with the traditional geodetic latitude, longitude, height coordinates {( ( h} or their mapping plane equivalents {N E H} a reference ellipsoid is introduced. Cartesian {X Y Z} coordinates are then transformed into their geodetic {( ( h} equivalents. To connect the mathematical world of the Cartesian coordinates to the physical world of the gravity field, a geoid model is used to compute geoid - ellipsoid separations. The geoid is "an equipotential surface which coincides with mean sea level, and which may be imagined to extend throughout the continents.". Qatar's geoid is based on the OSU91a geoid model from Ohio State University. Preliminary use of the OSU91a geoid revealed a long wavelength slope in the model that was unacceptable. It was therefore necessary to determine the slope and remove it. Program FITPLAN was obtained from the Geodetic Survey of Canada (GSC) to compute and remove the slope. Extensive new leveling (960 km) was undertaken to obtain spirit leveled heights for all network stations. Second order leveling was performed using the NA3000 and NA3003 digital levels. All existing and new leveling observations were combined and a re-adjustment of the leveling network in Qatar was performed. Given the ellipsoid heights (h) from the horizontal network and the orthometric heights (H) from the vertical adjustment it was possible to compute geoid-ellipsoid separations N* using the formulae N* = h - H. Separations were computed at 72 network stations (the other 7 stations were eliminated from the computations due to poor or unavailable spirit leveled heights) and introduced as constraints into the final computation of the Qatar95 geoid model. Analysis shows that the absolute accuracy of the geoid model is about 7 cm (95% confidence level) and the relative accuracy of the geoid is about 2 ppm. 4.0 Qatar National Datum QND95 Underlying any coordinate used for Surveying & Mapping, or indeed any positional information, is a Geodetic Datum. It is imperative that the datum is well defined and reproducible so that any positioning activity may be related to it. Two datums are in use in the State of Qatar; World Geodetic System 1984 (WGS84) and Qatar National Datum 1995 (QND95). QND95 is the datum used for surveying, mapping and related activities while WGS84 is the datum realized through the use of surveying techniques involving the Global Positioning System (GPS). Coordinates derived through the use of GPS are immediately transformed to QND95 except in a few circumstances such as the Civil Aviation Authorities who use WGS84 to be compatible with their world-wide counterparts. Similar to most jurisdictions, the State of Qatar uses a map projection of the reference ellipsoid so that users may use Cartesian { N E H } coordinates rather than the more complex geodetic coordinates { ( ( H }. Two projections are used in Qatar; Qatar National Grid (QNG) and Universal Transverse Mercator (UTM). QNG is encouraged throughout the State and is the only projection supported on QND95. UTM is available on the WGS84 datum only. 4.1 WGS84 WGS84 used in Qatar is defined by the United States Defense Mapping Agency (now NIMA) in their document "Department of Defense World Geodetic System 1984 - Its Definition and Relationship with Local Geodetic Systems", DMA TR 8350.2, dated September 30 1987 [Second Printing]. 4.2 QND95 QND95 is defined by transforming WGS84 coordinates using a 7 parameter coordinate transformation, where the 7 parameters are: X - translation = delta X = +119.42480 m Y - translation = delta Y = +303.65872 m Z - translation = delta Z = +11.00061 m   X - rotation = rx = - 1.164298 arc-seconds Y - rotation = ry = - 0.174458 arc-seconds Z - rotation = rz = - 1.096259 arc-seconds Scale change = delta s = - 3.657065 ppm The 7 parameters, { delta x, delta y, delta z, rx, ry, rz } are exact values, that is, QND95 is defined by taking a WGS84 coordinate and applying the above transformation. 4.3 Map Projections 4.3.1 UTM The UTM map projection used in Qatar is the standard definition of UTM. It is only used on the WGS84 datum not QND95. All UTM values in Qatar fall in zone 39 (Central Meridian 51° East Longitude). 4.3.2 QNG QNG is a transverse Mercator map projection using the following values: Northing origin N 24° 27' 00" Easting origin E 51° 13' 00" False Northing 300,000.0 m False Easting 200,000.0 m Scale factor at origin 0.99999 (exact) Note that QNG is defined in terms of the scale factor at the origin (central meridian) not the zone width. 4.4 Coordinate Transformations QTRANS In order to facilitate transformations between the WGS84 and QND95 datums program QTRANS is available. It uses the 7 parameter transformation detailed above and recognizes the various coordinate types. QTRANS also contains the Qatar95 geoid model which it uses in all computations. Figure 1 is the screen that is presented to QTRANS users. Figure 1: QTRANS Interface 5.0 On-line GPS Data Creation of the GPS Base Station has important implications for on-going survey work. It ensures that any survey that uses the data from the Base Station will automatically be in the QND95 system (assuming that the Base Station is held fixed in an adjustment and the seven parameter transformation is applied to the resulting coordinates). This same theory applies to any of the other 78 stations in the primary network but the advantage of an extra receiver is lost. One must remember however, that the accuracy of the tie depends on the distance from the Base Station and the precision of the equipment the user has. For example, assume the user's equipment has a precision of 1 cm plus 2 ppm and you are 50 km from the Base Station, then the expected accuracy is 11.0 cm. Table 3 provides a summary of various accuracies in cm based on an instrument with a precision of 1 cm + x ppm, where x = 1, 2, 5, and 10 ppm. Distance from Base PPM km 1 2 5 10 5 1.5 2.0 3.5 6.0 10 2.0 3.0 6.0 11.0 20 3.0 5.0 11.0 21.0 50 6.0 11.0 26.0 51.0 100 11.0 21.0 51.0 101.0 150 16.0 31.0 76.0 151.0 Table 3 - Accuracy of ties to Base Station in cm To use the data from the Base Station the user needs to receive it. Three possibilities exist for transferring the data to the user: on floppy diskette; via bulletin board service (BBS); or broadcasting. Broadcasting will be discussed under real time GPS. Floppy diskette would be an acceptable method, however, it requires a great deal of time and effort by CGIS staff. A user would have to contact the Centre with the pertinent details, then the staff would have to extract the data and save it on a diskette, and finally it would have to be delivered. A BBS is a much more efficient method of transferring the data. In this scenario the Centre provides an on-line Bulletin Board and the user simply logs into the BBS on the telephone lines and extracts the data they need. It can be directly downloaded to their own computer. No communication with CGIS staff or effort on the their part is required. The BBS contains the 14 most recent days data on-line ready to be extracted. Older data is archived. Users are encouraged to extract their data as soon as possible so that it will not have to be retrieved from the archive. Code and Phase data on L1 and L2 in the RINEX v2.0 (Receiver INdepentant EXchange format) is available for 1 second epochs. Figure 2 is a schematic of the Base Station and the BBS. The BBS is currently being replaced by an Internet Web Site. Figure 2 - Bulletin Board Service 5.1 Real Time Differential GPS GPS as discussed previously in this paper involves post processing of the GPS observations. That is, observations are first made in the field and then brought back to the office where they are combined with concurrent observations at the Base Station to determine the base line vectors. Post processing provides the highest accuracy available from GPS. Most traditional surveying and mapping applications, such as control surveys, mapping control, and GIS data collection fall into this category. Many applications however, require real time positioning. Real time positioning implies that the position (coordinates of the GPS receiver) are required directly in the field. Figure 3 is a schematic of real time positioning. Figure 3 can be contrasted to Figure 2 which shows a post processing scenario. Figure 3 - Real Time Positioning 5.2 Real Time GPS Theory Instantaneous positions determined by GPS receivers are limited to an accuracy of approximately 100 metres with Selective Availability turned on. For many applications, such as en-route navigation this may be sufficient, but for others, such as harbour navigation or emergency vehicle routing, 100 metres is not sufficient. The answer lies in the establishment of a Base Station at a known location. At this station it is possible to determine a "measured range" (pseudorange) to any particular satellite using the transmitted satellite signals (codes) and at the same time determine a "calculated range" because the coordinates of the satellite are known from the satellite orbit and the coordinates of the Base Station are known. The difference between the measured range and the calculated range is called the "differential correction". Differential corrections are calculated for all visible satellites at the Base Station. For distances of approximately 500 km away from the Base Station, these differential corrections may be considered equal for each satellite. If the roving receiver has access to them, then the differential correction can be added to the measured pseudoranges before calculating the coordinates. Differential corrections are distributed by broadcasting them either through the air (Low, Medium, High, Very High frequency microwaves or FM radio waves), via land lines, or via satellite communications. By far the most common methods of broadcasting are through the air and by communication satellites. Satellite communications require very large and expensive receivers so are often used in marine applications such as ocean navigation. On land, LF, HF, VHF and FM radio waves are the most common methods. FM radio broadcasts are popular in many parts of the world because of the abundance of FM radio stations and the inexpensive communication devices such as pagers for reception of the broadcast signals. In Qatar however, the service is not available so a VHF system was installed. VHF communication is limited to line-of-sight, typically 40 to 50 km depending on the height of the broadcast antenna and the terrain characteristics. LF and HF systems have much greater coverage but tend to be quite expensive and require special licensing. Expected accuracies of real time coordinates computed with differential corrections using VHF broadcasting are 1 metre, assuming the differential corrections are received within 6 seconds of broadcast. If the corrections are not received until 30 seconds from broadcast the accuracy is reduced to about 5 metres. Many new systems are on the market which use UHF communications and compute positions using the observed phase rather than the broadcast codes. These so called "real time kinematic" systems have centimetre accuracy but are limited in range to around 10 km. They also require special software at the base station (not the same Base Station as set up in Doha) and corresponding UHF radios at all GPS sights. They are most commonly used for GIS data collection and layout surveys. 6.0 GPS Validation Networks It is common practice for a country to set up calibration networks or baselines for its Surveying and Mapping measurement equipment. Most common networks are for Electronic Distance Measuring (EDM) equipment, precise leveling equipment, angle measurement equipment, and photogrammetric blocks. Similarly, networks are established for validation of GPS equipment and procedures. Note that GPS networks are called validation networks rather than calibration networks because they are not used to "calibrate" the equipment but to "validate" the observation and computation procedures. The main purposes for the GPS validation networks are to: Verify GPS equipment, Verify user observing procedures, Verify user computation procedures, and Ensure the combination of user procedures and computations produce results (coordinate and variance covariance information) that are compatible with those obtained by the country's survey authorities. In designing and establishing a GPS validation network, it is imperative that the different GPS survey methodologies, which have been designed to take advantage of the GPS equipment's characteristics and to maximize the results of the survey, be considered. The common survey methodologies and their predominant characteristics are given in Table 4. In the table the predominant characteristic is whether the methodology is usually used to survey points or lines (i.e. trajectory of the moving receiver). Methodology Predominant Characteristic Static Points - long or short distances Rapid Static Points - short distances Pseudostatic Points - long or short distances Stop and Go Many points - short distances Kinematic Trajectory - points Real Time Kinematic Trajectory - points On the Fly Kinematic Trajectory - points Navigation with Differential Corrections Trajectory - points Navigation Trajectory - points Table 4 - Survey Methodologies From Table 4 it is obvious that two different validation networks are required, one for long distances that is "point" oriented and the second that is "point" oriented for shorter distances and is also capable of testing "trajectories". Two networks have thus been established in Qatar, a "Static Validation Network" and a "Kinematic / Stop and Go Validation Network". 6.1 Validation Procedures Two distinct steps or phases are required in the establishment and use of a validation network. Firstly the network is designed and established, including an on-going observation plan and secondly a series of user procedures are established. The phases are briefly described below. 6.1.1 Phase I - Design and Establish Network Design the network considering the type of GPS survey to be validated. Major design considerations are: Monument stability, Line lengths, Station access, and Station suitability (i.e. sky visibility, multipath) Make the initial observation of the network. Longer than normal observation periods (usually twice as long) are chosen and all combinations of stations are observed for network strength. Compute the network to establish the network coordinates and their variance covariance matrix. If any doubt exists at all in the analysis of the observations repeat the session. Establish a re-observation schedule and computation procedures. Normal procedures are: Re-observe at least once a year, Compute new coordinates and variance covariance matrix, Statistically test results to ensure monument stability, Statistically test results for coordinate shifts, and Determine "new" network coordinates and covariance matrix. If monuments are stable and coordinate shifts have not taken place then re-compute the network using all observations gathered to date. Otherwise, determine why changes have taken place then compute a new set of coordinates and variance covariance matrix for the network based on the best observations available. In this case it is wise to re-observe the network a second time to verify the decisions taken. 6.1.2 Phase II - User Procedures User observes the network using their own observation procedures. User computes the network coordinates and variance covariance matrix using their own software and procedures. Statistically test the results to ensure that the validation network values are obtained. If statistically significant differences occur then determine if they are a result of observation or computation procedures. This is a difficult procedure and usually requires a great deal of co-operation between the user, the concerned authorities (Centre for GIS), and sometimes even the instrument manufacturer and/or software provider. As an option, user observations may be used by the establishing agency to "strengthen" the validation network. This should be done only when it is certain that the observations are fully compatible with those previously collected and will enhance the validation network. 7.0 GPS Education and Training The Centre for GIS considers training to be a key fundamental in the successful implementation of a GIS. Training in GPS is no exception. Various programs have been developed and offered to CGIS staff, other governmental staff, and the general public. The courses include: A one day "Management Overview" of GPS. A two day course which consists of the "Management Overview" plus in-depth treatment of more advanced theory including "GPS Co-ordinate Systems", "GPS Survey Methods, Planning and Field Surveys", "CGIS Base Station", and "GPS Economics". A one week course which includes further technical information on "GPS Measurements and Signals", "GPS Data Reduction and Analysis", plus hands-on field experience with GPS. The courses have been given several times with overwhelming response. 8.0 Conclusions GPS has proven to be a very valuable tool at the Centre for GIS, firstly to define the underlying coordinate system for the GIS, and secondly as a tool for gathering geo-referenced data. The geodetic network has ably supported new applications and has proven very reliable. GPS Base Station data is used regularly and its use is increasing as users become aware of its existence. Moving the BBS to the Internet will also increase its visibility. The QTRANS software has been sold both inside and outside Qatar and is used extensively. The Centre for GIS provides GPS training on an ongoing basis. At the Centre it is strongly believed that to properly use and implement GPS in any a thorough approach to its implementation is necessary. Consideration of geodetic datums, geodetic networks, geoid models, map projections, data collection and dissemination, and training are all mandatory for success. | Abstract Of The Paper & The Profile of The Speaker | Speaker Index | Paper Title Index | CGIS HOME PAGE CONTENTS