Trends in GPS Technology & Applications Chris Rizos Satellite Navigation & Positioning Group, School of Surveying and Spatial Information Systems The University of New South Wales, Sydney NSW 2052, AUSTRALIA Tel: +61-2-9385 4205, Fax: +61-2-9313 7493, E-mail:
[email protected]
ABSTRACT Since the launch of the first Global Positioning System (GPS) satellite 25 years ago, the technology and applications of satellite-based positioning has developed rapidly. There is no sign of this trend waning, and in fact we are on the verge of a massive explosion of new markets for GPS and associated position determination technology. GPS is but the 1st generation Global Navigation Satellite System (GNSS). By the end of the decade GPS will lose its ‘monopoly’, and will be but one component of a GNSS that includes the EU’s planned Galileo system (and possibly the Russian Federation’s revitalised GLONASS). Nevertheless, the amazing positioning capability that GPS has given the civilian and military communities, at no direct cost to the user, has to be acknowledged. Plans for the modernization of GPS are well advanced, and will enable for the first time a truly multi-frequency GPS. Around the same time as the GPS modernization phase is coming to conclusion (at the end of the present decade) Galileo will come online as a 2nd generation GNSS, with a variety of possible signals and services (some fee-based). Space-based and ground-based augmentation systems will continue to be deployed to address various GNSS user requirements for increased accuracy, signal availability and system integrity. Already accuracy augmentation has been available since the early 1980s, through the use of relative or differential GPS positioning techniques, instead of the standard point positioning technique. Very accurate (cm-level) positioning is now routinely possible using carrier phase-based techniques, even in real-time and for moving receivers. Current standalone GPS positioning is almost at the one metre-level of accuracy. The GPS user equipment has undergone a dramatic transformation from very heavy, expensive, analogue-based electronics to tiny digital devices that can be embedded within mobilephones. While professional applications of GPS (geodesy, surveying, mapping, navigation) have continued to benefit from GPS’s global availability, it is expected that GNSS will be increasingly used to support consumer applications for ‘personal and vehicular location’. This paper gives a brief overview of GPS technology and application trends.
1. INTRODUCTION 1.1 Background Since the launch of the first Global Positioning System (GPS) satellite 25 years ago, the technology and applications of satellite-based positioning has inextricably progressed. With every technological innovation, a new class of positioning applications was addressed. There is no sign that this trend is waning, and in fact we may be on the verge of a massive explosion of new markets for GPS and associated position determination technology. GPS is the first (and currently only) operational Global Navigation Satellite System (GNSS). The term “GNSS” was coined in the 1990s to acknowledge that GPS is merely the 1st generation of satellite-based positioning technology, and that by the end of the current decade there would be one (or more) independent GNSS that would complement and/or compete
with GPS. Nevertheless, many of the positioning and technological concepts, as well as the current and future markets and applications of GNSS, can be analysed using GPS as the exemplar. Although GPS was designed from the start to be a dual-use technology (although the military use was considered more important), it was not until the mid-1990s, when GPS was declared operational, that the number of civilian applications and volume of civilian GPS receivers grew rapidly. It is worth recalling the past history of GPS development, in order to compare it with the predicted future for GNSS in general. Although a significant generalisation, we can characterise the 1980s as the years when GPS was primarily of interest to the military and to specialist research applications such as Geodesy. During these years the satellite constellation was not yet complete (there was no real-time, ‘24/7’ operation), and hence civilian users were those who were content with collecting GPS measurements, and then processing these measurements in offline (post-mission) mode. It was at this time that the first significant civilian innovation was made, the development of high accuracy relative positioning techniques based on the analysis of carrier phase data. An excellent review of the utility of GPS for geodetic applications is Evans et al. (2002). Many of these geodetic applications are still important today. During the 1990s, and particularly after GPS was declared fully operational (with a minimum of 24 orbiting satellites), the benefits of GPS for marine and air navigation became obvious. It was during the 1990s that international organisations such as the International Civil Aviation Authority (ICAO) and the International Maritime Organisation (IMO) adopted new navigation concepts based upon the global and ubiquitous positioning capabilities of GPS. The market for land navigation systems was also launched. This was also the decade of Selective Availability (SA), the intentional degradation of the GPS positioning capability by the system operators. This had two main effects, to encourage the development of differential GPS techniques (to ‘claw back’ the accuracy ‘lost’ due to SA), and to strengthen the arguments made by many countries that the control of GPS by the U.S. military authorities was ultimately not in the best interests of the world community. The 1990s also saw first the full deployment of the Russian Federation’s GLONASS satellite constellation, and its subsequent slow demise by the end of the 1990s. Nevertheless, GLONASS was the first nonU.S. GNSS to challenge the monopoly of GPS. At this time, early in the first decade of the 21st Century, GNSS is poised to experience an explosive growth, as it is increasingly embedded within consumer electronics devices and is used to address the needs of a new mass market in ‘Telematics’-type products and services. Within the next few years it is predicted that as much as 90% of the market for GNSS products and services will be for personal and vehicle applications. Although it is sometimes difficult to distinguish between so-called Location Based Services (LBS) and Transport Telematics, in this paper a distinction will be made between ‘personal location’ (PERLOC) and ‘transport telematics’ (TRANTEL) when considering the market development drivers, and the challenges (technological and economic), for such GNSS applications. This paper gives a brief overview of the trends in GNSS technology and applications, dealing with the following matters: • GPS and the next generation GNSSs (section 2). • The augmentation of GNSS signals by space-based and ground-based transmitters (section 3). • Receiver technology and developments (section 4).
•
User applications and markets (section 5).
The analysis presented in this paper does not claim to be exhaustive. The author has taken a very personal approach and highlighted developments that, in his opinion, are significant in understanding how the future of GNSS will play out. The focus is on what we know about the GNSS technology, and its expected development, at the dawn of this century. In this paper sufficient background on GPS is given so that the predicted scenarios for GNSS technology development and applications can be better appreciated. The following sections provide a summary of the GPS system, its signals, and the user capabilities currently available. 1.2 Introduction to GPS The NAVSTAR Global Positioning System (GPS) is a satellite-based, all-weather, continuous, global radionavigation and time-transfer system, designed, financed, deployed and operated by the U.S. Department of Defense. The concept of GPS was initiated in 1973, and the first GPS satellite was launched in 1978. In 1993 the system was declared fully operational. GPS technology was designed with the following primary objectives: • Suitability for all classes of platforms (aircraft, ship, land-based and space), and a wide variety of dynamics. • Real-time positioning, velocity and time determination capability. • Availability of the positioning results on a single global geodetic datum. • Restricting the highest accuracy to a certain class of users (military). • Redundancy provisions to ensure the survivability of the system. • Providing the service to an unlimited number of users worldwide. • Low-cost, and low power users’ unit. GPS consists of three fundamental segments: Satellite Segment, i.e., the satellite constellation itself, the User Segment, including all GPS receivers used in a variety of civilian and military applications, and the Control Segment, responsible for maintaining the proper operability of the system. The Control Segment consists of five monitor stations, each checking the exact altitude, position, speed, and overall health of the orbiting satellites 24 hours a day. Based on these observations, the satellite orbital position and clock bias, drift and drift-rate can be predicted for each satellite, and then transmitted to the satellite for re-transmission back to the users via a navigation message modulated on the downlink signals. The nominal GPS constellation consists of 24 satellites (although there are currently 29 deployed) that orbit the Earth at an altitude of ~20,000 km, in just less than 12 hours (Figure 1). The satellites approximately repeat the same track and configuration once a day, advancing by roughly 4 minutes each day. They are placed in six nearly circular orbital planes, inclined at about 55 degrees with respect to the equatorial plane, with nominally four satellites in each plane. This configuration assures the simultaneous visibility of five to eight satellites (sometimes more) at any point on Earth. Since February 22, 1978 – the launch of the first GPS Block I satellite – the system has evolved through several spacecraft designs, focussed primarily on the increased design life, extended operation time without contact from the Control System (autonomous operation), and better satellite clocks. Block II satellites are the first operational satellites (first launched in February 1989), Block IIA satellites are the second series of operational satellites (first launched in November 1990), and Block IIR satellites, the operational replenishment satellites have carried the GPS into the 21st century (first launch in January 1997). The Block IIF, the follow-on satellites, are expected to have
their first launch in 2006. Information about the current status of the constellation can be found, for example, at http://www.navcen.uscg.gov/gps/. The signal generated by a GPS satellite ‘clock’ consists of three components: (1) pure sinusoidal waves or carriers (L1=154¥10.23MHz, L2=120¥10.23MHz), (2) pseudo-random noise (PRN) codes, and (3) the navigation message (Figure 2). There are two PRN codes, the precise P(Y)-code, superimposed on the L1 and L2 carriers, and the so-called coarseacquisition C/A-code, superimposed on the L1 carrier. All signals transmitted by GPS satellites are coherently derived from a basic frequency of 10.23MHz, as shown in Table 1.
Figure 1. The GPS constellation.
Code Modulation and Navigation Message L1 carrier f1 = 1575.42 MHz l1ª 19 cm
lC ª 293 m lP ª 29.3 m
L2 carrier f2 = 1227.60 MHz l2ª 24 cm
lP ª 29.3 m
C/A code (SPS) P (Y) code (PPS)
P (Y) code (PPS)
Figure 2. GPS signal characteristics. Table 1. Basic components of the GPS satellite signal. Ratio of Component Frequency fundamental [MHz] frequency fo Fundamental frequency fo 10.23 1 L1 Carrier 1575.42 154fo L2 Carrier 1227.60 120fo P-code 10.23 1 C/A code 1.023 fo/10 Navigation message fo/204600 50⋅10-6
Wavelength [cm] 2932.6 19.04 24.45 2932.6 29326 N/A
Each satellite transmits a unique C/A-code (which is in fact the satellite’s designated ID) and a unique one-week long segment of P-code. The PRN is a very complicated digital binary
code, or a complicated sequence of ‘on’ and ‘off’ pulses that looks almost like random electrical noise. This carrier modulation enables the measurement of the signal travel time between the satellite and the receiver (user), which is a fundamental GPS observable. Access to the C/A-code is provided to all users, supporting the so-called Standard Positioning Service (SPS). Under the Anti-Spoofing (AS) policy imposed by the U.S., the P-code is encrypted to form the Y-code, available exclusively to the military users, for a service known as the Precise Positioning Service (PPS). PPS guarantees positioning accuracy of at least 22m (95%) horizontally (and 27.7m vertically), while guaranteed positioning accuracy of SPS before Selective Availability was turned off 2 May 2000 was 100m (95%) horizontally (and 156m vertically). Under AS, the civilian receivers must use special signal tracking techniques to recover the observable on L2, since no C/A-code is available on L2. This will change when GPS is ‘modernized’ (section 2.1). 1.3 Positioning with GPS There are two fundamental GPS measurements: pseudorange and carrier phase, both subject to measurement errors of systematic and random nature. For example, systematic errors due to ionosphere or troposphere delay the GPS signal, and cause the measured range (or distance) to be different from the true range by some systematic amount. Other errors, such as the receiver noise, are considered random. Pseudorange is the geometric range between the transmitter and the receiver, distorted by the lack of synchronisation between the satellite and the receiver clocks, and the propagation media (atmosphere). It is recovered from the measured time difference between the epoch of the signal transmission and the epoch of its reception by the receiver. The actual time measurement is performed with the use of the PRN code. In principle, the receiver and the satellite generate the same PRN sequence. The arriving signal is delayed with respect to the replica generated by the receiver, as it travels ~20,000km. In order to find how much the satellite’s signal is delayed, the receiver-replicated signal is delayed until it falls into synchronisation with the incoming signal. The amount by which the receiver’s version of the signal is delayed is equal to the travel time of the satellite’s version. The travel time, Dt (~0.07s), is converted to a range measurement by multiplying it by the speed of light. There are two types of pseudoranges: (1) C/A-code pseudorange, and (2) P(Y)code pseudorange. The precision of the pseudorange measurement is partly determined by the wavelength of the chip in the PRN code. Thus, the shorter the wavelength, the more precise the range measurement would be. Consequently, the P(Y)-code range measurement precision (noise) of 10-30cm is about 10 times higher than that of the C/A-code. The pseudorange is the principle observable used for conventional single receiver positioning (under the SPS or PPS). Carrier phase is defined as a difference between the phase of the incoming carrier signal and the phase of the reference signal generated by the receiver. Since at the initial epoch of the signal acquisition, the receiver can measure only a fractional phase, the carrier phase observable contains the initial unknown integer ambiguity. Integer ambiguity is a number of full phase cycles between the receiver and the satellite at the starting epoch, which remains constant as long as the signal tracking is continuous. After the initial epoch the receiver can count the number of integer cycles that are being tracked. Thus, the carrier phase observable can be expressed as a sum of the fractional part (measured with mm-level precision), and the integer number of cycles counted since the starting epoch. The initial ambiguity can be determined using special techniques referred to as ambiguity resolution techniques. Once the ambiguity is resolved, the carrier phase observable can be used to determine the user’s location to a high accuracy for geodetic and surveying applications.
The principle behind positioning with GPS is “trilateration” in space. Essentially, the problem can be described as follows: given the position vectors of GPS satellites (i.e. orbital information) tracked by a ground receiver, and given a set of range measurements to these satellites, determine the position of the user. A single range measurement to a satellite places the user somewhere on a sphere with a radius equal to the measured range. Three simultaneously measured ranges to three different satellites place the user on the intersection of three spheres, which corresponds to two points in space. One of them is usually an impossible solution that can be discarded by the receiver. Even though there are only three fundamental unknowns (3D coordinate of the user’s receiver), a minimum of four satellites must be simultaneously observed to provide a unique solution (Figure 3), because the lack of synchronisation of the user receiver and satellite clocks requires the receiver clock error to be also determined. X2, Y2, Z2 X3, Y3, Z3
P2r
P3r
X4, Y4, Z4
X1, Y1, Z1 P1
P4r r
Unknown location of receiver r
Figure 3. Determination of position in space by ranging to multiple satellites.
There are two primary GPS positioning modes: point positioning (or absolute positioning), and relative positioning. However, there are several different strategies for GPS data collection and processing, relevant to both positioning modes. In general, GPS can be used in static and kinematic modes, using either pseudorange or carrier phase data (or both). GPS data can be collected and then post-processed at a later time, or processed in real time, depending on the application and the accuracy requirements (see, e.g. http://www.navcen.uscg.gov/pubs/frp2001/default.htm). In point positioning a single receiver observes pseudoranges to multiple satellites to determine the user’s location (Figure 3). For the positioning of the moving receiver the number of unknowns per epoch equal to three receiver coordinates plus a receiver clock error (or correction – to give precise time) term. The satellite geometry (see later discussion on DOP) and any unmodelled biases will directly affect the accuracy of the absolute positioning. The relative positioning technique (also referred to as differential GPS, or DGPS) employs at least two receivers, a reference (or base) receiver, whose coordinates must be known, and the user’s receiver, whose coordinates can be determined relative to the reference receiver. Thus, the major objective of relative positioning is to estimate the 3D baseline vector between the reference receiver and the unknown location. Using the known coordinates of the reference receiver and the estimated DX, D Y and DZ baseline components, the user’s receiver
coordinates can be readily computed. The most important advantage of relative positioning is removal of systematic error sources from the observable, leading to increased positioning accuracy (especially when combined with the use of carrier phase data). Since for short to medium baselines (a few tens to hundreds of km) the systematic errors in GPS observables due to troposphere, ionosphere, satellite clock and broadcast ephemeris errors are of similar magnitude (i.e., they are spatially and temporally correlated), DGPS allows for the removal, or at least significant mitigation, of these error sources. DGPS services have evolved during the 1990s, when SA was turned on, and are now commonly provided by the government, industry and professional organisations, and enable the user to use only one GPS receiver collecting pseudorange data, while still achieving superior accuracy as compared to the point-positioning mode. Naturally, in order to use a DGPS service, the user must be equipped with an additional hardware capable of receiving and processing the differential corrections. In an alternative implementation of DGPS, the reference station broadcasts the raw measurements (can be both carrier phase and pseudorange) instead of the differential corrections. DGPS services normally involve some type of wireless transmission system. They may employ VHF or UHF systems for short ranges, low-frequency transmitters for medium ranges (beacons) and L-band or C-band geostationary satellites for coverage of entire continents. Wide Area DGPS (WADGPS) involves multiple GPS base stations that track all the GPS satellites in view and, based on their precisely known locations and satellite broadcast orbital information, estimate the errors in the GPS pseudoranges. This information is used to generate pseudorange corrections that are subsequently sent to the master control station, which uploads checked and weighted corrections to the communication geostationary satellite, which transmit the corrections to the users. The positioning accuracy of WADGPS is typically at the sub-metre level. Other types of free-to-air DGPS services in North America include the FAA-supported (Federal Aviation Administration) satellite-based Wide Area Augmentation System (WAAS – section 3.1), ground-based DGPS services (section 3.3), referred to as Local Area DGPS (LADGPS), such as the U.S. Coast Guard and Canadian Coast Guard services, or the FAAsupported Local Area Augmentation System (LAAS – section 3.3). LADGPS supports realtime positioning typically over distances of up to a few hundred kilometres, using corrections generated by a single reference station. WAAS is currently operational, and LAAS is under implementation, with a major objective of both augmentation systems being support for civil air navigation. Another strategy gaining popularity in many countries is to establish local networks of continuously operating reference stations that can support a range of applications, especially those requiring the highest accuracy in post-processing or in real-time (section 3.4). Scientific organisations such as the International GPS Service deploy and operate global networks, allowing the users free access to the archived data (http://igscb.jpl.nasa.gov/). Alternatively, network-based positioning using carrier-phase observations with a single user receiver in realtime can be accomplished with local specialised networks (Rizos, 2002b). Depending on the design of the GPS receiver and the measurement type/technique employed, the positioning accuracy with pseudoranges varies from 10m with SA turned off (about 100m when SA was turned on during the 1990s), to better than 1 centimetre, in the case of carrier phase-based relative positioning. The positioning accuracy of GPS depends on several factors,
among them are the number and the geometry of the observed satellites, the mode of observation (point vs. relative positioning), type of measurement used (pseudorange or carrier phase), the measurement model used, and the level of biases and errors affecting the measurements. The geometric factor, Geometric Dilution of Precision (GDOP), reflects the instantaneous geometry related to a single point. In general, more satellites yield a smaller DOP value, and a GDOP of six or less typically indicates good geometry. Other DOP quantities, such as Position DOP, or Vertical DOP or Relative DOP (related to the satellite geometry with respect to a baseline), can be also used as quality indicators. Other factors affecting the GPS positioning accuracy depend on: (1) whether the user is stationary or moving (static vs. kinematic mode), (2) whether the positioning is performed in real-time or offline, (3) the data reduction algorithm used, (4) the degree of redundancy in the solution, and (5) the measurement noise level. The currently achievable GPS accuracies, at the 95% confidence level, are summarised in Table 2. The lower bound of the relative positioning accuracy cannot be stated with precision as it depends on several hardware and environmental factors, as well as the satellite geometry, among others. Thus, the horizontal accuracy levels listed in Table 2 should be understood as the best achievable accuracy; the symbol Æ indicates the increase in magnitude of the values listed. Table 2. Currently achievable horizontal GPS accuracy (Rizos, 2002a). Positioning mode Point positioning with pseudoranges Relative positioning Static survey 2 mm (Æ) plus 1 ppm PPS 1-5 m carrier phase (up to < 0.1 ppm) SPS, SA off 4-10 m Kinematic 5 mm (Æ) Good sky view (low survey carrier DOP) phase 10 m Æ Urban environment &/or multipath-affected SPS, SA on 0-100 m DGPS, 50 cm (Æ) pseudorange ppm – parts per million
1.4 GPS Instrumentation Over the past two decades the civilian as well as military GPS instrumentation has evolved through several stages of design and implementation, focussed primarily on achieving an enhanced reliability of positioning and timing, modularisation and miniaturisation, and reduction in cost and power requirements. One of the most important aspects, especially for the civilian market, has been the decreased cost of the receivers, as the explosion of GPS applications calls for a variety of low-cost and application-oriented equipment. By far the majority of the receivers manufactured today are of the C/A-code single-frequency variety. However, for the high precision geodetic applications the dual-frequency receiver configuration is the standard. Section 4.1 summarises trends in GPS receiver development. Even though the civilian and military receivers, as well as application–oriented instruments, have evolved in different directions, one might pose the following question: Are all GPS receivers essentially the same, apart from functionality and user software? The general answer is, yes, all GPS receivers support essentially the same functionality blocks, even if their implementation differs for different types of receivers. The following are the primary components of a generic GPS receiver (Figure 4): antenna and preamplifier, radio-frequency (RF) front-end section, a signal tracker block, microprocessor, control/interface unit, data
storage device, and power supply (Grejner-Brzezinska, 2002). Any GPS receiver must carry out the following tasks: • Select the satellites to be tracked based on calculations of its likely visibility. • Search and acquire each of the GPS satellite signals selected. • Recover navigation data for every satellite. • Track the satellites, measure pseudorange and/or carrier phase. • Provide position/velocity information. • Accept user commands and display results via control unit or a mobile device. • Record the data for post-processing (optional). • Transmit the data to another receiver via radio modem for real-time DGPS solutions (optional).
Multiple channels
Antenna and Preamplifier
Control and Interface unit Code tracking loop RF
Microprocessor
Carrier tracking loop
Data Storage
Power Supply Unit
Figure 4. Basic components of a GPS receiver (Grejner-Brzezinska, 2002).
1.5 Shortcomings of GPS GPS has, over the last two decades, enjoyed tremendous success, and it is now a ‘first choice’ position, velocity and time (PVT) technology for new civilian and military applications. While civilian use far outweighs military usage (it is estimated that there are 100 times the number of civilian GPS receivers vis-à-vis military receivers), GPS remains a curious ‘child of the Cold War’. The effectiveness of GPS-guided munitions in the recent Afghanistan and Iraq conflicts was a dramatic demonstration of the tactical and strategic power of precise positioning technology. This will ensure that GPS remains firmly under the control of the U.S. Department of Defense, and that future GNSSs will have built-in military functionality. Because the GPS signals are offered free to all users, across the world, and the receiver technology is comparatively mature, reliable and inexpensive (at least the instrumentation for single-frequency SPS-type positioning), GPS’s popularity is rapidly rising. Perhaps the fastest growing class of applications are the ones we refer to here as PERLOC and TRANTEL (section 5). However, GPS has many shortcomings, but none greater than that arising from the weak strength of the signal at the receiver antenna. The received L1 signal strength is of the order of –160dB (10-16 Watt), and the signal can be easily blocked by buildings and other objects, including foliage. Standard GPS therefore cannot be used indoors or in urban
environments where there are many signal obstructions. Although this has permitted other location technologies to gain some popularity (section 4.3), it has also spurred GPS R&D into weak signal acquisition and tracking (section 4.2). Furthermore, the GPS signals can also be easily jammed or degraded by unintentional RF interference. This vulnerability in terms of signal availability has been identified in a report by Volpe (2001). In addition, there is no efficient means by which users are warned when the satellite system is not operating within specifications, i.e. GPS’s integrity cannot be assured. Hence, for critical user applications such as air and marine navigation, GPS-only positioning is not recommended, and must be augmented in some way (sections 3.1, 3.2 & 3.3). GPS shortcomings (some of which are also relevant for all GNSSs) can be summarised as follows: • Weak signal strengths make the signals vulnerable to interference (intentional or unintentional), exposing critical user applications to possible episodes of denial-ofservice. • GPS in its standard mode cannot be used indoors, under trees, or in urban environments that may reduce the availability of navigation signals. • There is no built-in integrity warning/assurance signal or service for GPS users. • In order to improve the accuracy of GPS to the metre-level or below, relative or differential GPS techniques must be used, leading to more complex (and more expensive) user technology. • Only single-frequency GPS positioning is possible using standard GPS receivers, with expensive dual-frequency GPS receivers justified only for geodetic, survey and scientific applications. • Civilian and military applications both use the same GPS signals. • GPS control is firmly in the hands of the U.S. military (although there is some indirect U.S. civilian input via the Interagency GPS Executive Board, http://www.igeb.gov). • The budget and maintenance of GPS is subject to an annual tussle in the U.S. Congress (though it is embedded within the military budget). • GPS signals do not have a communication capability, hence requiring the integration of separate wireless comms technology for many Telematics-type applications (sections 5.2 & 5.3). 2. THE FUTURE OF GNSS By the end of the first decade of the 21st Century there will be several GNSSs, each with unique features, each operated by different agencies, and each ‘chasing’ the same user markets. We may speak of generations 1 (GPS and GLONASS), 1.5 (modernized GPS) and 2 (Galileo and GPS-III), operating side-by-side over the next 10 or so years. The best outcome would be that although the various GNSSs would be independent of each other (so that one catastrophic failure would not bring the whole GNSS down), they would also be compatible and interoperable so that user equipment may be developed that can utilise some or all of the broadcast signals (Figure 5). Let’s look briefly at the future GNSS components. 2.1 GPS Modernization and GPS-III GPS modernization refers to the collection of system improvements (satellite, signal, and control segment) that will change GPS from a 1st generation GNSS to what might be called a 1.5G GNSS, principally by offering additional user signals. GPS modernization is focussed on
improving the accuracy and addressing signal vulnerability for civilian uses of GPS, primarily through the implementation of a new PRN ranging code on the L2 signal, and a new civilian signal at the L5 frequency of 1227.6MHz (Spilker & Van Direndonck, 2001). The former permits civilian users to make pseudorange measurements on either L1 or L2 or both, providing system redundancy. An open (non-encrypted) L2 signal will lead to a reduction in the cost of dual-frequency receivers. The latter (L5) signal is in the protected aeronautical frequency band (Table 3), and is intended to satisfy civil aviation and other safety-of-life applications. The combination of three frequencies will revolutionise carrier phase-based position techniques, as ambiguity resolution will become a comparatively simple and robust accuracy enhancement process. Note that these new signals will still be provided free to all users. The new dual-frequency civilian tracking capability will be available on the Block IIR GPS satellites scheduled for launch starting 2004, while the other improvements (including the transmission of L5) are intended for Block IIF satellites that are scheduled for launch starting in 2006. However, it may be as long as 10 years before a full ‘modernized’ constellation is available. In the meantime hybrid GPS positioning will be the norm, with receivers having to track both the ‘old’ GPS satellites, as well as the ‘modernized’ satellites progressively brought online. In addition, a new (encrypted) M-code will be implemented exclusively for military use, ensuring that military and civilian users will have entirely separate signals and codes. Consequently, both the SA and AS policies will have finally been abandoned (SA has already been turned off in 2000). However, it is unlikely that control of GPS will be wrested from the U.S. military. Studies for what we might call the 2nd generation GPS (the modernization referred to above is essentially an enhancement of the current GPS to raise it to a 1.5G GNSS), or GPS-III, have commenced. However, GPS-III satellites are unlikely to be launched before 2012, several years after Galileo is scheduled to be fully operational. 2.2 GLONASS GLONASS is a GNSS developed by the former U.S.S.R. during the Cold War, and like GPS was initially intended primarily for military applications. Although the signal and code structure is different from that of GPS (see, e.g. Seeber, 2003, and Table 3) there are also many similarities. For example, GLONASS is a dual-frequency satellite-based radiolocation system that permits pseudorange and carrier phase measurements to be made. In addition, both point positioning and relative positioning is possible, with similar levels of accuracy to GPS. GLONASS became operational with a full 24-satellite constellation in 1996. However, as of September 2003 the number of operational satellites has dropped to just eight; thus the system’s long-term stability is questionable. Recently the President of the Russian Federation has made a statement committing Russia to having again an operational GLONASS by the end of the decade. Nevertheless, there exist GPS/GLONASS receivers that take advantage of the extended constellation created by additional satellites in view. For more information on GLONASS the reader is referred to such web sites as http://www.glonass-center.ru/.
2.3 Galileo Since the late 1990s the European Union has been promoting the development of an independent GNSS under ‘civilian control’. The primary motivation has been to address the needs of the transportation sector (particularly civil aviation) for a GNSS with guaranteed integrity. Galileo is the concept that has now been approved for development and deployment by 2008. The signal and code structure is far more complex than for GPS, consisting of up to ten trackable signals and codes (Figure 5, Hein et al., 2003). There are common signals with GPS L1 and L5, permitting a significant degree of ‘interoperability’ (i.e. integrated GPS/Galileo receivers able to track signals from both constellations). Unfortunately there are no plans for a Galileo signal that overlays the GPS L2 signal. Increasing the number of signals-in-space (from nearly 60 orbiting satellites) is to be applauded, as this leads to greater availability, particularly in urban environments. The combination of GPS and Galileo would certainly benefit many users (except perhaps for those using very low cost receivers), see Table 3. Nevertheless, global and regional augmentation services would still be expected to address concerns for availability, integrity and accuracy of GPS, GLONASS and Galileo (sections 3.1 & 3.2).
Figure 5. Radio frequencies for GNSS: GPS, GLONASS and Galileo.
GPS has two free-to-air services: the SPS (civilian) and the PPS (military). However, Galileo proposes to offer four services, two free-to-air (equivalent to the GPS-SPS, and a safety-oflife service), one commercial (high accuracy, high integrity service based on several frequencies), and an encrypted service for public authorities such as police, etc. Galileo may therefore be described as a 2nd generation GNSS. Furthermore, it will be deployed several years before the first GPS-III satellites will be launched. For more information on Galileo the reader is referred to the magazine Galileo’s World, and the web sites http://www.galileopgm.org/index.htm and http://www.genesis.org/.
Table 3. A comparison of GPS and Galileo (USION, 2003).
2.4 Summary Remarks There is a bright future for GNSS. The monopoly that GPS currently enjoys will be challenged by the end of the decade by the EU’s Galileo system. A revitalised GLONASS may also be broadcasting signals. There is still considerable concern in the U.S. about the manner in which Galileo has been promoted, its planned signal structure, and how it will (unfairly?) compete with GPS (see, e.g., http://www.usatoday.com/money/industries/technology/2003-07-30-gps_x.htm). Be that as it may, the rest of the world welcomes more signals-in-space, and more user equipment options. Furthermore, it could be argued that without the EU’s plan for Galileo the U.S. government may not have been convinced of the need to accelerate the upgrade of the ailing GPS. However, all GNSSs suffer from essentially the same shortcoming, and that is the relatively weak received signal strength. Various augmentation systems, as well as other location technologies, will still be necessary in order to address most user concerns about accuracy, availability and integrity, as discussed in the next few sections. 3. THE AUGMENTATION OF GNSS During the 1990s GPS was recognised as not being accurate enough, on its own, for many mapping and navigation applications due to the imposition of Selective Availability (Table 2). Although ever since the 1980s the geodetic and surveying community was using relative positioning techniques, the ground-based reference receivers necessary to support such high accuracy applications merely recorded data that was subsequently processed offline. However, the demand for real-time dekametre (or better) positioning accuracy required an augmentation of GPS that was ultimately to also address the concerns of the lack of integrity of the GPS itself. Differential GPS (DGPS) is a form of GPS augmentation. Although the removal of Selective Availability on 2 May 2000 renders DGPS unnecessary for some applications, DGPS is still a requirement for any application where an accuracy consistently better than five metres is required, and where reliability and the other factors matter. This is because the total effect of
atmospheric and orbit biases on single receiver positioning is still at the few metre level. Since these errors are spatially (and temporally) correlated, DGPS is effective in reducing them. Differential services imply real-time services, and hence users operating a wide range of GPS receiver equipment must be able to receive and process DGPS transmissions. The Radio Technical Committee for Maritime Services (RTCM) Special Committee 104 has drafted a standard format for the correction messages necessary to ensure an open real-time DGPS system (http://www.rtcm.org/). Assuming the necessary infrastructure for a DGPS service is in place (reference receivers, communication links, and transmission protocols), the most important system parameters are: • Predictable accuracy: The accuracy of a system’s position solution with respect to a mapped point. Note that both the position solution and the map must be based on the same geodetic datum. • Repeatable accuracy: The accuracy with which a user can return to a position whose coordinate has been measured at a previous time with the same system • Relative accuracy: The accuracy with which a user can measure position relative to that of another user of the same navigation system at the same time. • Availability: The percentage of time that the services of the system are usable. • Coverage: The surface area in which the signals are adequate to permit the user to determine position to a specified level of accuracy. • Reliability: The probability that the system will perform its function within defined performance limits for a specified period of time under given operating conditions. • Integrity: The ability of the system to provide timely warnings to users when the system should not be used. The needs and requirements for DGPS services, in relation to various air, marine and land navigation applications are defined in the U.S. Federal Radionavigation Plan (RP) (http://www.navcen.uscg.gov/pubs/frp2001/default.htm), and the reader is referred to this publication for background information (although several other RPs are under preparation, the U.S. RP is the most comprehensive and accessible document). DGPS services can be found in many countries, primarily because the majority of system developers and service providers are global companies operating across different geographic regions. Some are commercial systems, and others broadcast messages ‘free-to-air’. Several of them are mature systems, while others will expand as additional reference stations are established, and new technologies and user markets are developed. 3.1 Current Space-Based Augmentation DGPS Systems In the case of commercial Wide Area DGPS (WADGPS) services, satellite augmentation implies that communication satellites are used to provide the data link between user and service provider. However, future satellite-based augmentation systems plan to have additional navigation signals transmitted by satellites other than those of the GPS constellation. Both of these classes of services are briefly reviewed. There are currently two global WADGPS services, one offered by two multinational companies Fugro and Thales. Such WADGPS services rely on a large number of groundbased reference stations around the world. The operations of sub-networks in different regions are coordinated from a small number of central base stations. The reference stations within a particular sub-network send their data by landline (or other means) to the base station, where
the correction data is generated and passed to an upload station that sends this data to a geostationary satellite. The satellite acts as a “bent pipe”, and transmits these corrections to users. In the early days of the service the Inmarsat satellites were used (for which a user required a gimbal-mounted directional antenna). Nowadays L-band mobile satellite communication systems are used in many parts of the world, which permit users to receive the message transmission via small modem devices with short whip antennas. These WADGPS services are strictly commercial operations offering several levels of service. In their simplest configuration they operate as Local Area DGPS (RTCM messages from one reference station), which can deliver a few metres accuracy up to a few hundred kilometres from a reference station, with accuracy degradation being a function of the distance from the nearest reference station. The WADGPS option uses a proprietary format to combine the correction messages from several reference stations, delivering sub-metre accuracy. Users must have a special decoder in order to convert the proprietary WADGPS message into RTCM format before it is passed to the serial port of their GPS receiver. The principle advantage of such services is that they are available anywhere in the world. The Wide Area Augmentation System (WAAS) is a Space-Based Augmentation System (SBAS) has been deployed by the U.S. Federal Aviation Administration (FAA) to support aviation navigation for the enroute through to precision approach phases of flight (http://gps.faa.gov/Programs/WAAS/waas.htm). Its components are shown in Figure 6. The satellite-based component consists of up to four geostationary satellites that will have three major functions: (1) the transmission of GPS differential corrections to users, (2) the transmission of integrity messages to users, and (3) the provision of a ranging capability (extra GPS-like signals). The geostationary satellite transmission will be on L1 using a C/Acode for ranging and a uniquely structured message to broadcast differential corrections and integrity information to users. Reception of the differential corrections from the geostationary satellites is subject to the line-of-sight condition, similar to the GPS signal. This means that the SBAS messages come into the receiver through the RF “front-end” section. WAAS therefore improves availability, reliability and integrity to aeronautical users (although other users also benefit). The ground-based segment includes some 25 Wide Area Reference Stations, two Wide Area Master Stations, two Ground Earth Stations to uplink the correction/integrity messages to the geostationary satellites. The specified accuracy of the system is 7.6m, 95% of the time. However, the actual accuracy is significantly better than this, being of the order of a metre in the service area of the continental U.S., Alaska and the Hawaiian Islands. GPS receivers have to be ‘WAAS-ready’ (via a software upgrade only) in order to decode and use the SBAS messages. However, WAAS is a public service and anyone equipped with a suitably modified receiver will have access to it. This means that for applications requiring metre-level positioning accuracy WAAS may become a viable alternative to commercial WADGPS services. Other countries or continents are also deploying similar augmentation systems. The European Geostationary Navigation Overlay System (EGNOS) and the Japanese Multi-Function Satellite-Based Satellite Augmentation System (MSAS) are examples of these.
Figure 6. U.S. Federal Aviation Administration’s WAAS.
3.2 Future Space-Based Augmentation Systems While the proposed Galileo system will incorporate both an integrity service and a DGPS (commercial) service, extra signals from other satellites would improve availability even further. Hence, even with the prospect of a combined (interoperable) GPS/Galileo-based GNSS by the end of this decade, there are several proposals to launch regional SBASs. These extra satellites would optimise the coverage of GNSS over a region by transmitting GPS-type and/or Galileo-type ranging signals. In addition, the transmissions would support other services, such as navigational integrity, remote sensing and broadband communications. The Quasi-Zenith Satellite System (QZSS) has been proposed by a consortium of Japanese companies to their government (Petrovski et al., 2003). A constellation of between three and seven satellites will transmit GPS-like signals (possibly also Galileo-like signals), to improve signal availability in Japan. Studies have shown that the combination of high latitude location and dense urban environment significantly reduces the GPS signal availability in Japanese cities. The QZSS would, in addition, provide broadband communication services. The QZSS concept is currently still under study, but the decision on the release of funding to permit the project to proceed is expected in 2004. Note that the coverage of the QZSS will improve GPS (and ultimately Galileo) signal availability for the entire North/South-east Asian hemisphere (including Australia), as indicated in Figure 7 for some of the orbit configuration options being investigated. Over the last few years the Chinese government has been deploying several geostationary satellites of their Beidou system, with the latest (the third) launched in early 2003 (see, e.g., http://www.space.com/missionlaunches/china_launch_030525.html). This system appears to be of a local design, possibly intended for future military applications, and not as an augmentation GPS or Galileo. However, recent announcements indicate that the Chinese government is interested in a collaborative Galileo project, although it is unclear whether they are proposing to assist with the development of the core space-ground segment program, or merely wishing to build several Chinese-owned Galileo-like satellites (e.g. http://www.atimes.com/atimes/Global_Economy/EE02Dj01.html). The latter would represent a SBAS. India has also announced plans for a SBAS known as GAGAN (Geo-Aided
Navigation) to augment GPS/GLONASS/Galileo in its region via several Indian geostationary satellites (http://www.hinduonnet.com/2002/11/29/stories/2002112905730900.htm).
Figure 7. Orbit options for Japan’s QZSS (Petrovski et al., 2003).
3.3 Ground-Based Augmentation Systems As with the SBASs described above, Ground-Based Augmentation Systems (GBASs) come in several varieties. There are, for example, GBAS designed exclusively for aviation users, that augment GNSS accuracy, availability and integrity, and those that are also (or primarily) intended for general users. As far as augmenting accuracy is concerned all are of the Local Area DGPS (LADGPS) variety. We will refer particularly to the LADGPS services established worldwide to aid maritime users, to the FAA’s Local Area Augmentation System (LAAS), and to AirServices Australia’s Ground-based Augmentation System (GRAS), both intended to aid air navigation. Maritime DGPS systems are a worldwide endeavour with national services being set up and maintained by the maritime authority of each country. Both the International Marine Organization (IMO) and the International Association of Lighthouse Authorities (IALA) have produced specifications and guidelines for marine DGPS systems (IALA, 1999). In the United States and Canada, the U.S. Coast Guard and Canadian Coast Guard are responsible for these services. This service utilises the marine beacon frequencies of 285-325kHz. The advantages of this frequency band are its availability and the fact that ground waves propagate long distances (several hundreds of km). Over land the range is less and depends on soil conductivity, however it is still of the order of a few hundreds of kilometres, which is well over the radio horizon. The service, being very reliable and available in the coverage areas, has been well received by users well beyond the maritime community, particularly by farmers using DGPS-guided vehicles for ‘precision agriculture’. As a result, it is being extended to cover the entire continental U.S. under the Nationwide DGPS (NDGPS) Service project (http://www.navcen.uscg.gov/dgps/NDGPS). The U.S. FAA plans to deploy LAAS to support precision approach and landing Category II/III applications (http://gps.faa.gov/Programs/LAAS/laas.htm). The system will be installed at selected airports and will service a limited area around these airports. A local ground system will generate differential corrections for the GPS, WAAS geostationary satellites and airport pseudolites (when used). A VHF transmitter will then be used to broadcast the GBAS
correction and integrity messages to local users, which are input to the GPS receivers via their serial port. Pseudolites (from “pseudo-satellites”) are a GBAS technology that provides similar capabilities to the signals from WAAS geostationary satellites. That is, provides extra ranging signals, as well as permits correction/integrity messages to be sent to user equipment. WAAS is a very expensive aeronautical navigation augmentation that few countries will be able to afford. Furthermore, many countries would be reluctant to surrender sovereignty over navigation in their airspace to the U.S. (WAAS), Europe (EGNOS or Galileo) or Japan (MSAS or QZSS). An innovative hybrid concept has been proposed by AirServices Australia. Their GRAS concept is an alternative to the FAA’s WAAS (section 3.1) in that it will support enroute and precision airport approach applications, but will use GBAS messages from ground-based VHF transmitters instead of SBAS messages transmitted by geostationary satellites (Crosby et al., 2000). As with WAAS, a ground network of reference receivers tracks the GPS satellites, and a master station computes the SBAS messages. These messages are distributed by landline to a network of VHF transmitters co-located with GPS receivers which convert the SBAS messages into GBAS messages before transmission to overflying aircraft (Figure 8). Hence a country can continue to control the ground infrastructure necessary to support air navigation services, without subscribing to global SBAS-type services over which they have no influence.
Figure 8. AirServices Australia’s GRAS concept (Crosby et al., 2000).
3.4 Continuously Operating Reference Stations There is a trend in many countries for the establishment of continuously operating reference station networks (CORS) to support a range of applications. These networks are generally operated by government organisations as a free service (though there are some receivers run by private companies and from whom data may be purchased). By using CORS data, users are, in principle, able to perform high precision positioning using a single GPS receiver,
hence halving equipment costs. Both pseudorange-based DGPS and carrier phase-based positioning relative to the CORS network stations is possible. Such CORS networks can be easily enhanced to offer real-time DGPS services, and there is little doubt that ultimately many CORS networks will evolve into multi-functional infrastructure to support the full range of user positioning needs in their area of influence. The International GPS Service (IGS) coordinates the operation of a global network of CORS (http://igscb.jpl.nasa.gov). IGS data is primarily used to support global and regional geodetic and scientific applications of the highest accuracy. CORS networks have also been established to address local geodynamic applications. The most impressive example is the Geographical Survey Institute's (GSI) GEONET in Japan, established in the aftermath of the 1995 Kobe earthquake. The GEONET consists of over one thousand GPS receivers deployed across the country, linked to the main archive centre in GSI's Tsukuba headquarters (http://mekira.gsi.go.jp/ENGLISH/index.html). Another example is the Southern California Integrated GPS Network (SCIGN), comprising over one hundred GPS reference stations jointly operated by several academic and government organisations (http://www.scign.org/). CORS networks have also been established in many countries simply to support high precision geospatial applications. It is not possible to list here the national CORS networks currently operational (or planned). For example, the U.S. CORS Network (Figure 9) is a cooperative program coordinated by the National Geodetic Survey (http://www.ngs.noaa.gov/CORS/cors-data.html). This network consists of hundreds of GPS reference receivers, operated by a variety of government, academic, commercial and private organisations, and includes some reference stations of the U.S. Nationwide DGPS network (section 3.3).
Figure 9. Some stations of the U.S. CORS network.
Such CORS network stations typically record dual-frequency GPS data at 15 or 30 second intervals (though there is now a trend towards one second data logging), and make the available the raw data to users as RINEX files. The IGS and local CORS networks can support web-based processing services such as A U S P O S
(http://www.ga.gov.au/nmd/geodesy/sgc/wwwgps/). AUSPOS (and other similar web services) requires the user to upload their data to the web site, subsequently sending the coordinate results to the user. However, increasingly CORS networks are supporting real-time carrier phase-based positioning via the transmission of raw carrier phase data and special ‘network corrections’. Network-RTK is a centimetre-accuracy, real-time, carrier phase-based positioning technique capable of operating over inter-receiver distances up to many tens of kilometres with equivalent performance to current single-base Real-Time Kinematic systems (Rizos, 2002b). Network-RTK requires a data processing ‘engine’ with the capability to resolve the integer ambiguities between the static reference receivers that make up the CORS network. The ‘engine’ must be capable of handling double-differenced data from receivers 50-100km apart, operate in real-time, instantaneously for all visible satellites. The Network-RTK correction messages can then be generated. In addition to the data processing engine, the Network-RTK system needs to have a data management system and a data communication system. It needs to manage corrections generated in real-time, the raw measurement data, ultra-rapid IGS orbits, etc. There are two aspects to the data communication system: (a) between the master control station (MCS - where the data processing engine and data archive are located) and the various reference stations, and (b) communication between the MCS and users. From the Network-RTK implementation point of view there are three possible architectures (Rizos, 2002b): (1) generation of a Virtual Reference Station and its corrections, (2) generation and broadcast of an Area Correction Model, or (3) broadcast the raw data from all of the reference stations. RTCM is currently developing data transmission formats to support network-RTK. Several commercial systems have been developed, and they have been deployed at locations around the world. 3.5 Summary Remarks LADGPS and WADGPS will continue to be required, although as the accuracy of single receiver positioning increases (due to GPS modernization and the impact of WAAS), DGPS techniques must evolve to satisfy sub-metre accuracy requirements. This is most easily done by switching from pure pseudorange-based techniques to those that also make use of the lower noise carrier phase data. We are already seeing commercial WADGPS systems offering services with a few decimetre accuracy. Commercial LADGPS systems will either die or evolve into carrier phase-based RTK systems, delivering sub-decimetre (and even centimetre) level accuracy. The IMO-supported LADGPS for maritime use will probably be the only pure pseudorange-based DGPS system that will continue to be operated on a global basis, free to all users in MF radio coverage areas. Real-time positioning services will increasingly be provided (free-to-air, or fee-for-service) by CORS as they upgrade in the coming years from passive data logging networks to active networks that transmit RTCM/RTK correction messages for a range of users. Even CORS networks designed primarily to support geodetic applications (such as the IGS, GEONET, and others) will soon be operating in real-time so that data is available to users (and service providers) without delay. The data from such CORS networks will therefore be used to provide services that compete with current commercial WADGPS services, as, for example, in the case of NAVCOM’s Starfire service (http://www.navcomtech.com/products.cfm). Augmentation of accuracy is of course only one type of augmentation. The aviation community is promoting augmentations that incorporate a high degree of integrity. There are
various SBAS or GBAS schemes, but all rely on a CORS network of receivers. This CORS network monitors the health of the GNSS signals, and transmits integrity messages to users via satcom or terrestrial wireless communication links. WAAS is the archtypical SBAS for aviation applications, whilst LAAS is the complementary GBAS. However, there are other system architectures possible, such as exemplified by the GRAS concept. Availability of signals-in-space is a serious concern. Apart from using extra ground-based transmission from ‘pseudolites’, the only means of doing this is to make sure that future GNSSs have signal structures that permit ‘interoperability’, for those users with the appropriate equipment. However, there are opportunities for countries or regions to develop SBASs that address the non-aviation users concerns of signal availability. The QZSS is one example of a SBAS that is intended to go beyond just satisfying the needs of safety-of-life applications, and is intended to augment not only GPS, but also the future Galileo GNSS as well. 4. GNSS RECEIVER TECHNOLOGY TRENDS There are many user requirements for positioning, and one would therefore expect a plethora of technological solutions. It is perhaps surprising that GPS has been so successful across a wide spectrum of applications. It is in fact the ‘first choice’ technology for all new civilian and military applications. However, it is still instructive to consider the positioning technology options from a ‘taxonometric’ perspective, as indicated in Figure 10.
Figure 10. Taxonomy of positioning systems (Drane & Rizos, 1998).
The most common positioning technologies are ‘wave-based’, relying on RF, ultrasonic or infrared wave propagation. The other main ‘branches’ are ‘dead-reckoning’ (inertial and orientation systems) and ‘signpost’ (beacon, proximity, RFID or cell-ID type systems). ‘Remote’ positioning refers to the circumstance where a network or centre computes position (as in a tracking-type application), as opposed to a ‘self’ positioning system where the position calculation is performed on the user’s device. The former may also be referred to as
“network-centric” positioning, while the latter is “user-centric” positioning (see section 4.3 for further explanation). GNSS is therefore a wave-based, satellite-self positioning system. The utility of GPS (and in future other GNSSs as well) varies from application to application. In some cases GPS has made such an impact that there is no competitive technology. Geodesy is an example of such an application area. In the case of high accuracy positioning in support of surveying and mapping operations, GPS is a very valuable technology that is but one part of the ‘toolkit’, and in some scenarios GPS is clearly the best option, but sometimes an alternative terrestrial-based system is used. For many navigation applications GPS-only positioning is not recommended. There has been a long history of integrating GPS with other navigation sensor technologies. For example, GPS integrated with an Inertial Navigation System (INS) has been the architecture of choice for marine and air navigation, machine guidance and control, field robotics, as well as for precise mobile mapping/imaging applications. INS developments are themselves moving very rapidly, with lower cost, higher performance, smaller sized MEMS (Micro-Electro-Mechanical Systems)-based sensors becoming increasingly an indispensable component of multi-sensor systems for a wide variety of civilian and military applications. The reader is referred to Grejner-Brzezinska (2001a,b) for a review on precise multi-sensor systems. The applications described above are all what might be referred to as professional applications. However, by far the majority of positioning applications will be associated with products and services marketed to consumers. We can identify two consumer application areas: ‘personal location’ (PERLOC – section 5.2) and ‘transport telematics’ (TRANTEL – section 5.3). PERLOC primarily is concerned with positioning a small portable device, and therefore GPS has to overcome its greatest shortcoming if it is to be used within mobile telephony technology. GPS (and GNSS in general) is not reliable, or accurate, when satellite signals are blocked and/or reflected, as is the case inside buildings or in ‘urban canyons’. Although significant progress has been made in improving weak signal tracking (section 4.2), it is still not clear whether GPS will be the only, or even the preferred, solution for mobilephone/personal-device position determination (section 4.3). Figure 11 illustrates how several positioning technologies could be used within an urban environment. TRANTEL applications, on the other hand, do have GNSS as its core positioning technology, though the incorporation of ‘dead-reckoning’ sensors does significantly improve availability and accuracy. Before discussing further the application trends for GNSS positioning, the following section summarises some of the GNSS receiver technology trends. 4.1 General Receiver Trends When considering the trends in GNSS receiver development we may parody the Olympian Ideal: ‘faster’, ‘smaller’, ‘cheaper’, and ‘better’. What do we mean by ‘better’? The answer is, lower power consumption, increased accuracy, less susceptibility to RF interference, multifeatured, improved APIs, and higher sensitivity to weak received signals (section 4.2). Modern digital receiver designs have achieved a high degree of miniturisation as a result of the microelectronics revolution. Current receiver designs are ‘postage-stamp’ size, consisting on two (and even one) hardware chips, making them well suited for embedding within more complex systems. Already over ten million GPS receiver chipsets have been embedded within mobilephones. In fact, we are already seeing the GPS receiver functions being subsumed, as
an IP-core, into a multi-purpose chip (see, e.g., http://www.centralitycom.com). In the near future full software-based receivers (based on DSP and FPGA boards) may reach similar levels of performance as current hardware-correlator designs. Such receivers will be easily reconfigureable, and hence will be upgradeable to track any GNSS (and/or augmentation system) signals.
Figure 11. Personal positioning technologies in urban environments (MIT, 2003).
The modernization of GPS and the deployment of Galileo over the next decade or so will have a profound impact on future GNSS receiver designs. Instead of single-frequency receivers being the norm (and expensive dual-frequency receivers used for mainly niche applications), multi-frequency receivers will be used for all but the most price-sensitive applications. Such receivers will permit higher positioning accuracy, will be less susceptible to RF interference, and will be able to acquire and track lower strength signals than current ‘standard’ GPS receivers. The cost of integrated GNSS receivers (e.g. GPS+Galileo, or GPS+GLONASS, or GPS+GLONASS+Galileo) will be of the order of 20-50% more than single-system receivers. In fact, the majority of the value of a receiver will be in its software (particularly application-specific software), not hardware. In such a short review paper it is not possible to present a comprehensive overview of GPS receiver developments (though the reader interested in tracking the evolution of GPS technology is referred to past issues of the GPS World magazine). What is presented below is therefore but a (subjective) summary of GPS receiver trends: • Small-sized, low-cost OEM receivers that are well suited for developing new applications. • Tracking of SBAS signals such as WAAS will be a standard capability of all receivers. • More sophisticated signal tracking, including multi-frequency, multi-GNSS capability.
• • • • • •
Greater degree of signal processing to mitigate multipath interference, and to track low strength signals. Increasing use of carrier phase-based receivers (especially with multi-frequency tracking capability) to support sub-metre positioning accuracy. Greater integration of GPS with other sensor technologies, and wireless communications. Enhancement of positioning capability through the ability to augment signals from terrestrial pseudolites (transmitting GPS ‘look-alike’ signals). More powerful onboard microprocessors, permitting the porting of more sophisticated application-specific software into the receiver’s firmware. The embedding of GPS receiver functions into multi-purpose chips.
4.2 Weak Signal Tracking & A-GPS The U.S. Federal Communication Commission’s (FCC) mandate to telecommunications carriers to deploy an ‘enhanced’ 911 emergency response system for mobilephones (known as ‘E911’ – http://www.fcc.gov/911/enhanced/) has been an important driver for the development of positioning technologies capable of being embedded within mobilephones. The E911 mandate does not specify which technology is to be used. It only defines the general specifications, including that the positioning accuracy of a device making a ‘911’ call be of the order of 50m (67% of the time) and 150m (95% of the time) for ‘user-centric solutions’, or 100m (67%) and 300m (95%) if a ‘network-centric solution’ is used. Initially it appeared that GPS would not satisfy such requirements, as GPS signals could not be received indoors. Hence much of the attention during the late 1990s has been on mobile telephonybased systems (section 4.3). It must be said, however, that although safety/security has been the main driver for mobilephone positioning in the U.S., in other countries it has been Location Based Services (LBS). Be that as it may, the number of GPS receivers currently embedded within mobilephones far outnumbers ‘standard’ GPS receivers. GNSS has enjoyed significant success as a position determination technology for almost all outdoor applications. However, its most significant shortcoming is the weak received signal strength, being of the order of –160dB Watts. During the last half decade or so there has been renewed interest in improving the GPS receiver technology in order to track very low signal strengths, as would be encountered inside buildings, of the order of 20-30dB below the ‘open sky’ signal strength. As a result, Wireless Assisted-GPS (A-GPS) and ‘high sensitivity’ GPS receiver designs have been developed (Djuknic & Richton, 2000), and such systems will increasingly be deployed within mobilephones, competing with (or complementing) other mobile telephony techniques. How can A-GPS help? For example, the timing and navigation data for GPS satellites may be provided by the network, which means that the receiver does not need to wait until the broadcast navigation message is read (even if it could decode this message, which may not be the case if the received signal is of very low strength). In essence, the assistance data makes it possible for the receiver to make the time measurements (equivalent to ranges) to GPS satellites without having to decode the actual GPS message, thus significantly speeds up the positioning process. A-GPS assistance messages (delivered to the GPS over the wireless telephony link) can also aid the receiver by providing information on visible satellites and their predicted Doppler-shifted signal frequency. Sophisticated new tracking signal processing algorithms have led to the development of so-called ‘high sensitivity’ receivers. Such receiver, with assistance data, would be able to acquire signals, make measurements and compute position almost instantly. The high sensitivity GPS receiver manufacturers include
Snaptrack (http://www.snaptrack.com), Global Locate (http://www.globallocate.com), SiRF (http://www.sirf.com), and Sigtec (http://www.signav.com.au). 4.3 Complementary & Competitive Mobilephone Positioning Technologies In this section we will focus on competitive and/or complementary location technologies to GNSS for PERLOC and TRANTEL applications (sections 5.2 & 5.3). These Telematics-type applications distinguish themselves from many other navigation/positioning applications in that what is required is a suitable relationship between Telematics Service Providers, mobile telephony networks and mobile users’ devices, in order to locate the user with a required accuracy and to deliver to him/her the appropriate location-filtered information. While ‘high sensitivity’ GPS is now a serious contender as a positioning technology for indoor and/or urban environments (section 4.2) it is not the only technology being considered for PERLOC, E911 or Location Based Services applications. Let us consider the options for positioning of a mobilephone, as this represents by far the largest potential market for PERLOC devices. The signal parameters most commonly used in mobilephone radionavigation are: angle-ofarrival, time-of-arrival, and signal strength (where the last two parameters correspond to the range measurement), and signal multipath signature matching. The most popular technique of finding the user’s location is ‘triangulation’, based on angular measurements, or ‘trilateration’ when distance measurements are used, or some combination of both, between the mobile user and the base stations. The base stations are either mobilephone service towers (cellular network) or GNSS satellites. Thus, in general, the technologies for finding the user’s location can be divided into network-based or satellite-based (currently GPS-based) systems. Another classification is based on the actual device that performs the positioning solution, i.e., mobile user or at some network control centre (NCC), leading to mobile terminal (user)-centric or network-centric, or hybrid solutions. In the network-centric systems, the user’s position is determined by the NCC and sent back to the user’s mobilephone, while in the terminal-centric solution, the position computation is performed by the user’s handset. In this section we present an overview of these three main PERLOC techniques. Much of this information has been taken from Hjelm (2002), Grejner-Brzezinska (2003), and web sites such as http://www.wirelessdevnet.com/channels/lbs/features/mobilepositioning.html. A summary of the characteristics of these techniques is presented in Table 4. The terminal-centric solutions rely on the positioning software installed within the mobilephone. They are further divided into: • GPS • Wireless-Assisted GPS (A-GPS, section 4.2) • Enhanced Observed Time-Difference (E-OTD) The GPS method provides instantaneous point-positioning information with an accuracy of 550m, depending on the availability of GPS signals (Table 2). A-GPS uses an assisting network of GPS receivers that can provide information over the mobile telephony network, enabling a significant reduction of the time-to-first-fix to 1-8s. The accuracy is also degraded relative to standard ‘open-sky’ GPS. Another terminal-centric solution is the E-OTD, which measures the time of the signal arrival from multiple base stations (within the wireless network) at the mobilephone. The time differences between the signal arrivals from different base stations are used to determine the user’s location with respect to the base stations. For the positioning and timing purposes, the base stations might be equipped with stationary GPS receivers. Thus, the base stations in E-OTD serve as reference points, similar to GPS satellites. However, this
method is not subject to limitations in signal availability affecting GPS, but still suffers from multipath (as all radionavigation systems do). The horizontal positioning accuracy of E-OTD has been quoted at about 100-125m (95% of the time). E-OTD can also be used in the network-centric mode. Table 4. Review of technologies for mobilephone positioning (Grejner-Brzezinska, 2003). Technique Primary Observable Upgrade of the User Location Calculation Terminal or Network & Control • Time (range) to multiple • User terminal (GPS GPS satellites receiver, memory, Mobile Terminal A-GPS • 3D location software) • Minimum of 3 ranges required • Non-synchronised for 2D positioning networks may require an enhancement Location/ • Multipath signature at the • None Multipath users location Network Pattern • 2D location Matching • Received signal strength • None RSS • 2D location Network • Signal travel time between the • Supports legacy terminals TOA user and the base stations • Monitoring equipment at Network • 2D location every base station • Signal travel time difference • User Terminal (memory, E-OTD between the user and the base software) Mobile Terminal stations • Base station time • 2D location synchronisation • Signal travel time difference • Network interconnection TDOA between the user and the base Network stations • 2D location • Time (range) to multiple cell• Network interconnection AOA towers (minimum of three • Antenna arrays to Network measurements is required) measure angles • 2D location • Cell ID • None CGI • The accuracy does not meet Network E911 requirements • 2D location Hybrid • GPS range • Same as for GPS method System • Cell ID Mobile terminal plus (A-GPS + • 3D or 2D Network CGI)
The main network-centric solutions are: • Cell Global Identity with Timing Advance (CGI-TA) • Time-of-Arrival (TOA) • Uplink Time-Difference-of-Arrival (TDOA) • Angle-of-Arrival (AOA) • Multipath (Location) Pattern Matching • Received Signal Strength (RSS) CGI uses the Cell-ID to locate the user within the cell, where the cell is defined as a coverage area of a base station (the tower nearest to the user). It is an inexpensive method, compatible with the existing devices, with the accuracy limited to the size of the cell, which may range from 10m to 500m (indoor micro-cell), to an outdoor macro-cell reaching many kilometres in
size. CGI is often supplemented by the TA (Timing Advance) information that provides the time between the start of a radio frame and the data burst. This enables the adjustment of a mobilephone’s transmit time to correctly align the time, at which its signal arrives at the base. These measurements can be used to determine the distance from the user to the base, further reducing the position error. TOA is based on the travel time (equivalent to distance) information between the base station and the mobilephone. In essence, the user’s location can be found by trilateration, at the intersection of three (or more) arcs centred at the tower locations, with radii equal to the measured distances, similar to single receiver GPS positioning (section 1.3). The concept of TDOA is similar to E-OTD, however, in TDOA the time of user’s signal arrival is measured by the network of base stations that observe the apparent arrival time differences (equivalent to distance-differences) between pairs of sites. Since each base station is usually at different distance from the caller, the signal arrives at the stations at slightly different times. To calculate the distance-difference between the two base stations, a hyperbola is defined, with each base station located at one of its foci. The intersection of the hyperbolas defined by different pairs of base stations determines the 2D location of the mobilephone. A minimum of three stations must receive the signal to enable the user’s location estimation as an intersection of two hyperbolas. The AOA method is based on the observation of the angle-of-signal arrival by at least two cell towers. The towers that receive the signals measure the direction of the signal (azimuth) and send this information to the AOA equipment, which determines the user’s location using basic trigonometry. The accuracy of AOA can be high, but may be limited by the signal interference and multipath, especially in urban areas. Much better and more reliable results are obtained by combining AOA with TOA. The Location (Multipath) Pattern Matching method uses multipath signature in the vicinity of the mobile user to find its location. The user’s terminal sends a signal that gets scattered by bouncing off the objects on its way to the cell tower. Thus, the cell tower receives a multipath signal and compares its signature with the multipath location database, which defines locations by their unique multipath characteristics. This method is comparatively unreliable and is unlikely to be favoured over other networkcentric methods. Another method is based on the signal strength model observed for the area. By merging the information about the actual Received Signal Strength (RSS) with an existing model, the system can predict the user’s location. An important feature of this technology is that it can determine the location of any wireless device with no modification or add-ons or enhancements to the mobile telephony network. The basic mathematical model is the relation between the signal strength and the distance between the mobilephone and the base. As in other ranging techniques, the user is located on a circle around the base with a radius equal to the distance measured using the signal strength. Accuracies of the order of tens to hundreds of metres have been reported. In summary, based on the time of the user’s signal arrival recorded at the base stations (TOA), or the time differences between the signal arrival from multiple base stations recorded by the mobilephone (E-OTD, CGI-TA), the user’s position can be determined by the standard method of “trilateration”, as the time observation can be converted to a distance measurement (or the distance-difference). Clearly, TOA or CGI-TA have the advantage over TDOA by working with existing GSM mobilephones, but may require significant investments in the supporting infrastructure (this is especially true with TOA). CGI-TA is rather inexpensive, as the cell information is already built into the networks. The E-OTD and TDOA methods require an extensive infrastructure support, moreover E-OTD needs also customised handsets.
It should be mentioned that the existence of a variety of positioning technologies, may pose a problem to both the users and the mobile telephony service providers, as the user is only covered in the area serviced by his/her provider and may not be covered elsewhere, if another provider uses a different location-identification method. It is rather difficult to define the ‘best’ technology, as each has advantages and disadvantages (Table 4). Perhaps a hybrid solution, for example, based on GPS with a network-based CGI-TA, would offer a more reliable solution. Other hybrid options are, for example, AOA plus TDOA (called Enhanced Forward Link Triangulation), E-OTD plus A-GPS, and AOA plus RSS. 4.4 Summary Remarks Although there are many different user requirements for positioning, over the last decade or so GPS has been very successful as a ‘first choice’ technology for all new civilian and military applications. In 2003 estimates of GPS receiver sales are of the order of US$15B, with the rate of increase over the last five years being 25% or so. Furthermore this rate of receiver sales increase is predicted to be maintained at the level of 20% pa through to the end of the decade and beyond. This is clearly good news for new GNSSs such as Galileo. GNSS is unlikely to be the sole positioning technology used for the majority of applications. Many precise navigation, machine control and mapping applications will use GNSS integrated with other navigation sensors, particularly of the ‘inertial’ variety, to ensure high accuracy, availability and integrity. However, the largest potential market for GNSS receivers is for personal and vehicle location applications. Although hybrid solutions will be offered, for the most price-sensitive applications there may only be one adopted positioning technology. In the case of vehicles GNSS is most likely to be the only (or at least the core) technology. For embedding within personal mobile devices it remains to be seen whether GNSS will be the technology of choice. 5. GNSS MARKET & APPLICATION TRENDS We may distinguish two broad classes of GNSS/positioning applications (or markets): (1) the professional applications, and (2) the consumer applications. All positioning applications appear to be growing at a respectable rate, with predictions of the annual value of GNSS products and services being several tens of billions of (U.S.) dollars within a few years. However, almost all GNSS market studies have suggested that the most dramatic increases will be in so-called ‘telematics’ products and services, which we identify here as being the market for personal location (PERLOC – section 5.2) and transport telematics (TRANTEL – section 5.3). The EU, in its study of the future of GNSS, has predicted that by the middle of the first decade of the 21st Century over 90% of the market for GNSS will be in personal location and transport applications. (Telematics is a recently adopted word referring to the convergence (or combination) of wireless telecommunications with (computer) information science/technology (IT). However, there is a certain ambiguity in the term ‘telematics’, as there are other application areas of increasing importance that also rely on IT and communication technologies. One of these is in the area of home-based Health Services, and the term Health Telematics is often used to refer to such applications as the remote monitoring of a person's physiological parameters, and response by health services when requested, telemedicine, and so on. A more accurate
descriptive term therefore is Transport Telematics. However, we believe that the essential driver for telematics is the increasing mobility of people, which in turn implies that a person’s location is a crucial attribute. Hence, to place greater emphasis on positioning within telematics some commentators have coined the word ‘telegeoinformatics’ (Karimi & Hammad, 2003).) 5.1 ‘Professional’ Application/Market Trends These range from the scientific applications, to engineering and navigation. The civilian applications, and some comments on trends, are summarised below: • Geodesy: GPS (and future GNSSs) will continue to support geoscientific studies into crustal deformation, climate change, vulcanology, seismology, oceanography and atmospheric monitoring, through the ongoing provision of products such as global tracking of GPS and other satellites, timing, precise orbits, tropospheric and ionospheric models, and the maintenance of the fundamental terrestrial reference frame. These activities are coordinated by the International GPS Service (http://igscb.jpl.nasa.gov/), and with the contribution of local elements such as national and regional CORS networks (section 3.4). • Surveying & Mapping: GPS is an important article of the surveyor’s ‘toolkit’, providing a capability for establishing high accuracy control for mobile mapping (from vehicles, airborne and marine platforms), deformation monitoring, and many civil engineering applications. The modernization of GPS (section 2.1), and the deployment of the Galileo system (section 2.3), will decrease the cost, and increase the flexibility, accuracy and robustness of carrier phase-based positioning, in turn promoting new markets for high precision positioning. • High Precision Kinematic Positioning: the trend to real-time centimetre-level positioning, supported by appropriate CORS infrastructure (section 3.4), and taking advantage of augmentations to the current GPS, will continue so that cm-level positioning will be as ‘easy’ as current metre-level positioning via pseudorange-based DGPS techniques. The markets for ‘precise navigation’, in such applications as field robotics, precision agriculture, and machine guidance and control, will steadily expand. However, to guarantee availability and integrity, GNSS may be integrated with other technologies such as INS and pseudolites, as well as signals from other SBAS (section 3.3). • Real-Time Navigation: GNSS has already revolutionised all navigation, on land, in the air and at sea. Despite concerns about its integrity, GPS has been readily accepted by the wider community. Only the regulated maritime and civil aviation transportation sectors have been resisting the inevitable. Several augmentation schemes of GNSS have been developed to improve the ‘raw’ accuracy, availability and integrity of GPS. These schemes include SBASs (for aviation systems such as WAAS) and GBAS (for aviation systems such as LAAS, and for other industry sectors the various LADGPS and WADGPS implementations). Hence modernized GPS, GLONASS and Galileo, together with SBASs and GBASs, will singly, or collectively, result in significant advances in marine and air navigation, leading to increased safety and efficiency. • Asset and Fleet Management: the ability to determine the position of fixed and mobile assets (be they buildings, pipelines and structures, vehicles, etc.) will lead to more effective utilisation of such assets. Prime examples of the management of mobile assets include ‘fleet
management’ of commercial vehicles (trucks, couriers, etc.), emergency vehicles (police, ambulance, firefighting, etc.), and public transportation (taxis, buses, ferries, trains). The tracking, and even guidance, of such vehicles is already a reality in many countries. With improvements in GNSS receiver and wireless comms technology on the one hand, and road map databases on the other, the market penetration of such applications will increase further. • Intelligent Transport Systems (ITS): ITS is a worldwide movement to improve the efficiency and safety of road transport, and safeguard the environment, primarily through an application of technology to reduce road congestion (Catling, 1994). One component of the ITS program is Advanced Traveller Information Systems (ATIS), and is related to the TRANTEL applications discussed in section 5.3 (Rizos & Drane, 2003). ATIS uses the position of a driver’s vehicle to aid his/her navigation (answering such questions as ‘where am I?’, ‘where is xxxx?’, ‘how do I navigate to yyyy?’). Other ITS applications use vehicle location information to better manage transport systems by, for example, monitoring traffic congestion and suggesting alternate routes for vehicles, or changing the phasing of traffic lights, and so on. An increasingly important ITS application will use vehicle location information provided by GNSS for electronic road pricing (ERP), such as in the case for trucks in Germany (see, e.g. http://www.transport-pricing.net/). ITS will therefore be a significant beneficiary of improved GNSS capability. • Innovative and Unusual Applications: Everyday new positioning applications of GNSS are identified. While some are trivial, but many contribute to the wellbeing of society. For example, there are many tracking applications, such as the monitoring of prisoners in the community so that they do not cross “geo-fences”, the tracking of endangered animals, the tracking of private or rental cars (for insurance purposes), and so on. Another huge application area comprises the geo-referencing of objects (or even events) for subsequent analysis within a Geographic Information System (GIS). These include geodemographic studies, geospatial market analysis, mapping of health patterns and the spread of disease, disaster management, etc. Augmented or mixed reality applications will also benefit from GNSS (and other) positioning technology (see also section 5.2). • Non-Positioning Applications: Today it is not fully appreciated that GPS plays a crucial role in defining and disseminating a stable and high precision timing standard. GPS timing is relied upon for the correct functioning of such critical infrastructures such as power grids, mobile telecommunication systems, computer networks, banking and finance (ATMs, stockmarkets, etc.), and the Internet itself. However, ground-based and satellite-borne GPS receivers are also increasingly used as atmospheric probing/monitoring tools, providing synoptic information on ionospheric activity and the wet component of the troposphere. In future we are likely to also see GPS used for indirect imaging of the ocean and ice surface, in bistatic radar modes of configuration. 5.2 ‘PERLOC’ Application/Market Trends PERLOC primarily is concerned with positioning a small personal device such as a mobilephone, ‘smartphone’ or Personal Digital Assistant (PDA). For the sake of this discussion we assume that the positioning technology is embedded within the mobile telephony device. PERLOC applications may require the user to determine their location anywhere, at anytime, but to the requisite positioning accuracy. Hence the greatest challenge is placed on positioning technology, as it must operate reliably indoors and outdoors. If
GNSS is to be used, it has to overcome its greatest shortcoming: GNSS (in general) is not available when satellite signals are blocked and/or reflected, as is the case inside buildings or in ‘urban canyons’. Although significant progress has been made in improving weak signal tracking (section 4.2), it is still not clear whether GPS will be the only, or even the preferred, solution for mobilephone/personal-device position determination (section 4.3). The driver for PERLOC technology and applications in the U.S. is the FCC’s ‘E911’ mandate (section 4.2), while in other countries it is largely Location Based Services (LBS). Nevertheless, once the positioning capability is established (no matter what the technology driver), it is likely that LBS will be an extremely important new market for mobile telephony service providers. What is a LBS and is it synonymous with PERLOC applications? LBSs can be broadly defined as those services delivered to a mobile device that use location as a ‘filter’ of irrelevant information, or as a trigger for a service, or to customise information. LBS typically are requested by a person (or at least not ‘blocked’ if LBS is ‘pushed’ to customers), and are very similar to the concierge services that make up the range of TRANTEL applications (section 5.3). Examples of LBS are: • ‘Finder’ services, e.g. ‘where is the nearest xxx?’ (xxx being a store, building, service, physical object, person, etc.). • Location-based marketing/advertising, where customers receive notifications of sales promotions if they are located within a store’s ‘catchment area’, e.g. the transmission of vouchers for discounted goods via SMS. • Navigation or route guidance instructions from Location ‘A’ to Location ‘B’. The crucial technologies are: (1) mobile positioning, (2) mobile telephony (or other wireless communications), (3) spatial databases of ‘points-of-interest’ (POI) and road networks, and (4) LBS servers that can handle millions of ‘transactions’ per day (e.g. as in the case of iMode - http://www.nttdocomo.co.jp/english/p_s/imode/, or J-phone services, in Japan). The LBS server is the oft-forgotten constituent technology, yet it is the ‘heart’ of PERLOC applications. In addition to managing the databases, for each LBS request (or transaction) the following operations may have to be executed: (a) “geo-coding” (convert a location, or address, into a position, or coordinate), (b) “reverse geo-coding” (convert a position/coordinate into a location/address), and (c) navigation from one location to another. For example, to find the nearest ATM, the customer’s position is transmitted to the LBS server. A spatial search is initiated within the POI database to locate the nearest ATM, and a step-by-step guide is sent to the person who requested the LBS (in the form of a customised map image, or a text message). LBSs are being trialled in many countries using available mobile telephony positioning techniques such as Cell-ID (section 4.3). The main challenges are the establishment of comprehensive road and POI databases, the development of realistic business plans, and the launch of compelling consumer services. Although the range of LBSs is limited only by one’s own imagination, LBSs must succeed in the marketplace and hence ‘cleverness’ is not enough. Nevertheless, in many respects the refinement of GNSS technology for PERLOC applications can proceed independently of the launch of specific LBSs. However, the development of more sophisticated LBS must await the maturation of accurate positioning technology (say at the 10-100m accuracy level) based on GNSS, but possibly augmented with other sensors (e.g. Figure 11). Location Based Services, although a convenient catch-all phrase, does not convey the full range of PERLOC applications. This author believes that the LBSs referred to above are just
the vanguard for many other location-enabled consumer applications, and to convey the full richness of potential PERLOC applications a new vocabulary may be needed. Some have suggested ‘L-commerce’, ‘context-computing’, ‘location-aware computing’, ‘geo-commerce’, ‘telematics’, and even ‘telegeoinformatics’ (Karimi & Hammad, 2003). Let us list some of the PERLOC applications not yet immediately identifiable with (current mobile telephony-based) LBSs: • Mobile Tourist Information Retrieval: Examples of these are electronic tourist or museum ‘guides’. Depending on the person’s position, different information may be provided. This information may be local (within the device), or retrieved from a server over a wireless comms link such as a WLAN (similar to the Server-assisted Electronic Navigation Assistant referred to in section 5.3). Already companies such as Lonely Planet have ‘electronic books’ or mobilephone services (http://www.lonelyplanet.com/mobile/), and adding ‘locationbrowsing’ capability, although not a trivial exercise (as it requires all content to be organised in spatial databases, as for LBSs), may be a future option. Certainly we will soon see implementations of such “e-guides” at outdoor tourist attractions such as archeological sites, and the like. Augmented reality technology may enhance the experience even further (see below). • Sport and Entertainment: Positioning technology will have three basic roles: (a) sporttraining, (b) participatory sport, and (c) spectator-sport. Obviously outdoor sports such as sailing, motor racing, golf, rowing, road bicycle racing, etc., are well suited for using GNSS technology (on its own or in combination with other sensors). Indoor positioning to the requisite accuracy and reliability is still not easily achieved (probably the best current technology uses CCTV cameras to triangulate ‘targets’ on the field), and therefore it is rare for field sports such as football, hockey, etc., to routinely use positioning technology. Being able to use position information (and velocity, acceleration, etc.) does contribute to sporttraining, as coaches can analyse the performance of athletes and/or equipment and correlate it with topography, location, and so on (http://www.dartfish.com). Knowing position can also assist golfers, as they will be able to derive information such as distance to the pin. Many other examples can be given including online gambling. However, perhaps the largest market for positioning is to enhance spectators enjoyment of sport. The technology of sport broadcasting is evolving rapidly, and future sport coverage will include spatial relationships of players, etc. • Geo-Games: This is perhaps the most inventive class of applications, referring to locationbased games. These may be of the ‘shoot-em-up’, ‘fugitive’ or ‘treasure hunt’ variety, or the more complex MMORG (Massive Multiplayer Online Role-playing Games). There are already GPS-based games (http://www.geocaching.com.au) and mobilephone-based games, and the distinction between video/computer games in future may become increasingly blurred (see web sites such as http://kmi.open.ac.uk/projects/presence/games_links/, to appreciate that ‘indoor’ and ‘outdoor’ games are now lumped together). As the positioning accuracy of ubiquitous mobile devices (such as mobilephones and PDAs) improves, and the mobile telephony bandwidth increases (as in 3G system), we can expect to see more sophisticated geo-games being developed. • Augmented/Mixed Reality: ‘Augmented’ or ‘mixed’ reality (AR) is arguably one of the ‘coolest’ applications of Location-Based Computing, and is actively researched in university computer science departments around the world. AR adds information (visual, oral, etc.) to ‘reality’ on the basis of the person’s location. Many examples of AR are
currently being researched or developed, in fields as diverse as sport, military training, construction engineering, electronic art, museum and archeological studies, simulationbased training, and even advertising! AR is a demanding application of all technologies: positioning, wireless communication, visualisation/rendering, database design and wearable computing. A search of the web will identify hundreds of interesting sites (see, e.g., http://www.augmented-reality.org/, and http:// www.csl.sony.co.jp/project/ar/ref.html). 5.3 ‘TRANTEL’ Application/Market Trends Intelligent Transport Systems (ITSs) are not so much a technology as they are a vision. In the 1980s, traffic planners and researchers saw that the application of ‘high technology’ offered the prospect of revolutionising their national transportation sector. They foresaw a transportation system of the future that would be more efficient, less polluting and safer (see, e.g. Catling, 1994). ITSs can be divided into five functional areas (Drane & Rizos, 1998), each briefly described in the following paragraphs: • Advanced Traffic Management Systems (ATMSs) involve the use of sophisticated technologies to manage the traffic on the transport network. An important element of ATMS is advanced traffic control systems that will, e.g., phase all the traffic lights in a particular area, provide such functionality as a “green wave” to vehicles. ATMS also includes electronic road pricing (ERP), adaptive signposting, traffic congestion monitoring, and incident management systems. • Advanced Traveller Information Systems (ATISs) are systems that provide information directly to the driver of a vehicle. An important service is route guidance, where the driver is informed of the best route to travel in order to reach a particular destination, taking into account road congestion conditions. ATIS also includes concierge services such as the location of nearby restaurants, parking space availability, and other geographically relevant information. • Advanced Vehicle Control Systems is the most ambitious of the ITS functional areas, and ultimately will involve having the vehicle controlled by computer, so that it can travel along the highway with no (or minimum) human intervention. In the short term this functional area deals with collision warning systems and intelligent cruise control. • Commercial Vehicle Operations involve the use of Automatic Vehicle Location Systems (AVLSs) linked with computer-aided dispatch systems. These systems enable more efficient dispatch and scheduling as well as increased driver safety. Many examples of such systems have already been deployed. These systems also have applicability to fleets of emergency vehicles, such as ambulances, police vehicles, etc., as well as public transport (section 5.1). • Advanced Public Transport Systems involve the development of special purpose public transport information and control systems. These will provide passengers with information on the arrival times of buses, trams or trains, allow smart card payment of fares, and a much higher level of operational efficiency. In addition, it will be possible to use ATMSs and ATISs to provide higher priority to buses and trams. This area is also likely to see the development of personalised public transport that will provide a service that is intermediate in terms of cost, timeliness, and proximity between those provided by buses and taxis.
These five functional areas give only a broad overview of ITSs, and many countries have adapted and implemented many of the constituent systems. The beneficiary or customer of an ITS system may be the vehicle driver, a passenger, a fleet manager, a dispatcher, a road traffic monitor, or a government regulator. In general, the task of such a system or service will be to answer a question that has a spatial aspect to it, such as: ‘where am I?’; ‘where are you?’; ‘where is the nearest parking station?’; ‘when is the next bus travelling south along Z Street?’; ‘what is the traffic density at Location X?’; ‘how many miles has the vehicle traveled since its last service?’; ‘which vehicle is nearest to Location Y?’; and so on. Each of these requirements or applications will involve a different configuration of the constituent technologies: IT devices and software, wireless telecommunications systems, position determination systems and spatially-referenced data. We wish to focus on the Transport Telematics (TRANTEL) applications that are essentially (though not exclusively) associated with the ATIS functional area (as exemplified by Figure 12). The following categories of TRANTEL applications are highlighted in this paper: • Driver Assistance: Systems that provide vehicle route guidance and navigation, enhance security and safety (such as calls for breakdown services and alerting authorities when an airbag is inflated), respond to requests for traffic information, concierge or Location Based Services (LBS), and so on. • Passenger Information: Products and services that target the passenger in an automobile (e.g. LBS, in-vehicle games and entertainment), or public transport (e.g. providing information on times of services, multi-modal connections, etc.). • Vehicle Management: Including such applications or systems as automated vehicle monitoring and fleet management, in order to improve the efficiency of commercial fleet operations, permit stolen vehicle recovery, monitoring of young drivers (by parents), remote opening of doors or disabling of a vehicle, the provision of real-time vehicle diagnostics services, and so on. The Vehicle Navigation System (VNS) is the quintessential in-vehicle terminal device. However, the sophistication of the VNS can vary considerably (Rizos & Drane, 2003): 1. The Electronic Street Directory (ESD) is its most rudimentary form and consists of a fixed in-vehicle LCD screen or removable Personal Digital Assistant (PDA), upon which map data is displayed. The map database is contained within a memory module (RAM, CD-ROM, compact flash, etc.). The system’s functionality is very limited, and generally restricted to ‘pan’ and ‘zoom’ of the digital map display, and as the name implies, merely replaces the book-style street atlas. 2. The Electronic Vehicle Locator (EVL) is an enhancement of the above, and permits the current vehicle's location determined by a positioning system to be displayed on the LCD screen. Such a combination of mobile map display and positioning system need not be permanently installed within the vehicle, and in fact the archetypal configuration is the PDA to which a GPS receiver is attached via the compact flash or serial port. The map database is identical to the one used in the ESD 3. The Electronic Navigation Assistant (ENA) makes use of navigable Digital Road Map (DRM) data to aid the driver. The DRM data can support either enhanced positioning via map-matching, or best-route calculation and route guidance. Typical installations involve an integrated GPS plus dead reckoning positioning system, an in-vehicle computer and LCD screen, a means of command input (mobile pointer/controller, or voice recognition),
and the DRM database (usually in the form of a CD-ROM or DVD-ROM). The DRM data include all geospatial information on roads and points-of-interest, and are able to support map queries such as ‘how do you get from my current location to location X?’, ‘where is the nearest petrol station?’, etc. 4. The Server-assisted ENA (SENA) is the logical extension of the ENA. The in-vehicle equipment is essentially the same as for the ENA, but by means of a wireless communications link the in-vehicle system can receive additional dynamic data, such as traffic congestion information, to improve best-route calculation and route guidance. In addition, this wireless link permits access to a wide range of LBSs and concierge services, significantly enhancing the driving experience.
Figure 12. Transport Telematics (TRANTEL) applications.
All VNSs except the ESD use GPS, on its own or in combination with other sensors. However, it is the SENA that is the defining example of a TRANTEL implementation that will gain wider acceptance than the other VNS forms referred to above, as the wireless connectivity permits access to a range of services: • Broadcast traffic information such as the level of current congestion on roads, reports of breakdowns or accidents likely to cause delays, highway construction information, traffic lights that are not functioning, locations of speed cameras, and so on. • Concierge services that find a particular type of restaurant, shop or point-of-interest (by accessing a database at a server), that reserve an item or service (e.g., a parking spot, a hotel room, etc.), or that otherwise enable a transaction to take place. • Request for roadside assistance in the event of a vehicle breakdown, the driver being locked out, or if an accident has occurred. • Alert emergency services, e.g., if there has been a car crash, or the vehicle has been stolen. • Route guidance in which driver instructions are generated by a central server and delivered in the form of voice commands. In fact the SENA concept, when carried to its natural conclusion, means that all of the services can be accessed from a very ‘thin’ mobile device, even a PDA or ‘smartphone’. Therefore, the SENA device need not even be permanently installed within a vehicle, and the various concierge and LBSs can be offered to all mobile users, whether in a vehicle or not. All they require is a mobile device with the necessary positioning and wireless communication link accessories, and hence the distinction between TRANTEL and PERLOC applications is blurred. It is possible to distinguish between two types of wireless connectivity: (a) the continuous broadcast of information to vehicles (TRANTEL), and (b) information sent upon request by a user (either TRANTEL or PERLOC). However, it must be emphasised that TRANTEL applications imply real-time services delivered via a wireless link. Where the wireless infrastructure is not established, or where the appropriate services cannot address users requests for assistance, then full advantage of in-vehicle Telematics technologies cannot be taken. Although the rate of new and after-sales installations of SENAs in vehicles varies from country to country, the unmistakeable trend is for an increase as more and more compelling TRANTEL services are offered to drivers. The main types of broadcast information relate in some way or the other to traffic conditions, e.g., breakdown incident alerts, weather and road traction conditions, and information on traffic congestion, road restrictions and parking availability. Such information has been broadcast to drivers by AM/FM radio for decades. However, the trend is to broadcast such information in a standardised digital format that computers can readily understand and use in, e.g., best-route calculations and route guidance. In Europe, the Radio Data System (RDS) is the delivery channel for real-time traffic and weather information via the Traffic Message Channel (TMC) (http://www.tmcforum.com). The TMC is a set of message formats that can be decoded by a TMC-equipped car radio or VNS. Although the current channel is based on RDS, TMC can also be delivered via digital radio, Internet, or GSM/GPRS systems. The Vehicle Information & Communication System (VICS) was established by the Japanese government, coming online in 1996, and currently provides real-time traffic and accident information at no charge to drivers. Drivers can receive VICS information through a variety of means, principally via special beacon receivers installed in vehicles or FM broadcasting. The situation in the U.S. as a whole is less well developed. To support such continuous broadcast services, a significant investment in road
network monitoring technology is required, something the industry is reluctant to undertake. Hence, much of the basic infrastructure to support broadcast services has been, to date, provided by government agencies. This trend is expected to continue for the foreseeable future. It is important to emphasise that the difference in levels of interest in broadcast road network information is not just a function of economic prosperity, but reflects the different priorities of the vehicle driving population. In the U.S., Driver Assistance Services to enhance safety and security are considered a higher priority than services that aid navigation on congested or unfamiliar road networks, as is the case in Europe and North Asia. On the other hand, information or services requested by a driver are best provided by a pointto-point wireless link, such as mobile telephony. The mobilephone (or an embedded modem that accesses the mobilephone network) is therefore the primary enabling technology for the request to, and delivery of information from central servers or call centres. For example, a driver may initiate a “mayday” call, either by a voice call to a human operator or by the press of a button (placing the emergency call via a computer-to-computer hook-up). Alternatively, the “mayday” call could be generated automatically in the event of airbag activation (and the coordinates of the vehicle sent in a short message burst). There are essentially two means of accessing Driver Assistance services, either on an ad hoc (e.g., once-off) basis, or via subscription to a service provider. The former effectively provides similar traffic information that the European and Japanese broadcast systems currently do, but does so over a telephone link. In the U.S., since mid-2000, the telephone number ‘511’ has been reserved for callers who are seeking information on traffic and weather conditions in their state (http://www.its.dot.gov/511/511.htm). This free service delivers the information to the caller via voice, but cannot provide input to an onboard navigation computer. The ‘511’ service provides similar information to web-based services currently available for many cities, counties and states around the world. In 1996, General Motors (GM) launched its OnStar system (http://www.onstar.com). Initially, the devices to access navigation and Driver Assistance Services were installed in top-of-theline vehicles. However, GM has expanded the range of vehicles it sells that have OnStar equipment factory-installed. Furthermore, several Japanese and German automakers have adopted the OnStar architecture. With over 2.5 million subscribers, OnStar is the most successful of the TRANTEL services offered to vehicle owners. Subscription costs vary from US$199 per annum for the basic service, rising to over US$300 for premium services. However, the setting up and maintenance of such services by automakers is a very expensive undertaking, and very few (if any) are profitable (the Ford-backed venture, known as Wingcast, was abandoned in mid-2002). The range of Driver Assistance Services include: air bag deployment notification, emergency services, roadside and accident assistance, stolen vehicle tracking, remote door unlock, remote diagnostics, concierge services, route support, and general information services. These services are accessed via a mobilephone link, transmitting voice requests and responses, or data, from the vehicle to the OnStar call centres/servers, and visa versa. Such subscription services can also be used by others apart from the drivers of private vehicle. The breadth of services and the variety of system configurations that support them is very wide, and the Passenger Information Services range from the frivolous to the critical, and cover a wide range of users: passengers within private vehicles, those engaged in pre-trip planning, and public transport commuters. The archetypal class of services are those that can be accessed by the passengers through the in-vehicle mobile terminal or via the mobilephone
(similar to PERLOC services). Increasingly, as the amount of time spent each day in traffic grows, efforts will be made to improve the quality or ‘productivity’ of this vehicle time. Although some services are available to the vehicle driver, given the potential for driver distraction, most of these services are offered to the other occupants. These services include access to electronic games, attending to email, receiving reports on weather, stock prices and sports scores, and ultimately (if there is enough bandwidth in the wireless communications link) downloads of music and video clips. Not all of these use the vehicle’s position as a ‘filter’, and hence not all can be considered examples of LBSs. However, the same IT and wireless communications components are used as in the case of Driver Assistance Services. Furthermore, extra infrastructure does not need to be deployed to allow for the provision of such services. The trend for the automobile becoming ‘wired to the Internet’, of being an entertainment centre, and even of evolving into a mobile office (or home) is unmistakable. Many organisations are seeing the vehicle (and its occupants) as a new source of revenue for services and products. All this is made possible by the revolution in auto electronics and wireless communication networks. The vehicle’s location simply enables the process by providing a means of filtering unwanted or irrelevant services. The Driver Assistance Services can be also accessed during pre-trip planning. ESD, ENA or SENA-type VNSs can be used to simulate the intended trip and to obtain, e.g., best route information before setting out. This would be done in the vehicle if the terminal is permanently installed within the vehicle. Hence, details, e.g., of the nearest petrol station to the trip’s destination could be obtained, and programmed into the in-vehicle device. Alternatively, if the mobile device is a PDA, then this pre-trip planning can occur anywhere, at home, in the office, and even on the move. In addition, web-based services could be accessed at home, or from any convenient PC or Internet-enabled device. Passengers planning to use public transport could obtain advice on their itinerary, with stepby-step instructions on which buses or trains to use. They could be informed of which transport services are running late, and the estimated arrival times of the service at stops of interest to the passenger. Currently, such public transport information can be obtained via the web, via information kiosks, the Internet, LCD screens in the bus or train, or computer displays at selected bus or train stops. Increasingly such services will be available to mobile devices such as mobilephones and PDAs, but in a more ‘intelligent’ manner than is currently the case. PDAs can be used to access the web via wireless links (such as mobile telephony, or WLAN-IEEE802.11 systems), but they are indistinguishable from other (wired) web users. If position information was automatically part of the query process, then the public transport information provided to the user could be tailored to the device’s location. Hence, only information on transport options relevant to that place and time needs to be sent. The infrastructure needed to support public transport Passenger Information Services is currently being deployed as part of programs to improve the efficiency and attractiveness of public transport. Hence, buses, trains, trams, etc., will need to be tracked using some form of remotepositioning technology, and this information used to predict the time the services arrive at different stops on their route. In many respects such operations are similar to commercial or emergency fleet management, with one major difference – public transport routes are fixed. The challenge therefore is primarily in distributing this information to passengers, or potential passengers. Accessing such information from mobile personal devices, via wireless communication networks, is a natural extension of current practice. Vehicle Management Services (VMS) are equally as broad as Passenger Information Services, but with the primary difference being that the services are not, in the first instance, targeted to
the driver or passenger of the vehicle. We can identify the following VMS: •
Commercial Vehicle Operations: Fleets of trucks, couriers, taxis and other commercial vehicles can be managed using TRANTEL applications such as the Fleet Management System (FMS). The ‘back-office’ FMS application is generally a form of a GIS, containing DRM data and incorporating data about the vehicle fleet, customer addresses and other information. The FMS has information transmitted to it of each vehicle's position, and often other data such as engine performance, as well as cargo-specific information such as the temperature of refrigerated goods, etc. If the goods are of a hazardous nature (e.g., waste, chemicals, dangerous materials, etc.), or valuable (e.g., cash, electronic goods, cigarettes or alcohol, etc.), then the FMS is effectively a means of tracking the progress of consignments. On the other hand, the FMS can be used to dispatch the appropriate vehicle to a customer, on the basis of the location of the vehicles in relation to the customer. Yet another use of the FMS is to help schedule vehicle maintenance. This class of logistics application is expected to continue to grow strongly.
•
Emergency Vehicle Dispatch: In many respects the operations of police, ambulance and fire fighting vehicles are similar to those of commercial vehicle fleets. Therefore, they can be managed by a FMS, and in general the FMSs for emergency vehicles are identical to those used for commercial vehicle operations. However, the most important functions of such FMSs are: dispatch of vehicles (including route guidance), monitoring of the vehicles and occupants (to ensure the safety of emergency personnel), managing situations (e.g., monitoring the nature of the emergency, and the response), and handling requests by emergency personnel (e.g., for additional or specialist services). Emergency vehicles typically have sophisticated communication systems, and the trend is to incorporate broadband wireless links.
•
Personal Vehicle Monitoring: A private vehicle may be tracked or monitored remotely for different reasons. The driver may request this monitoring (e.g., as a Driver Assistance Service if they are in an unfamiliar area and need guidance), or it may be initiated by the vehicle owner (e.g. in the case of a parent lending their car to a teenager driver), or it may be precipitated in the event of car theft or accident. Clearly, unlike Commercial Vehicle Operations, or Emergency Vehicle Dispatch, the monitoring of private vehicles is an ‘invasion’ of privacy, and hence such an action must be agreed to (explicitly or implicitly) by the owner or driver. Such monitoring may be passive, involving tracking only, or it may also lead to an action such as the switching off of the car's motor remotely, as in the case of vehicle theft.
•
Remote Vehicle Diagnostics: This is perhaps one of the more interesting of this class of TRANTEL applications. Imagine that the vehicle’s Engine Management System was monitoring the parameters of the vehicle, such as engine and brake performance, electronics, cooling and lubrication sub-systems, and transmitted this information to a service centre if something was on the brink of malfunctioning. The service centre could communicate with the vehicle, diagnosing the possible problem, alerting the driver to this, and even make an appointment to have it repaired! Such Remote Vehicle Diagnostic Services are already being offered for some luxury cars, and the trend will see such services extended to more and more vehicle owners.
All of these services require a data link from the vehicle to the dispatch or control centre, by which position information, driver requests, vehicle data, etc., are transmitted. This link may
be radio-based, as in the case of emergency and commercial vehicles, or provided by the mobile telephony network in the case of private vehicle services. The ‘tracking’ of the vehicle by an organisation is invariably required, raising the issue of privacy. Many commentators believe that this class of TRANTEL applications will be the most effective, as they impact directly on industrial productivity, as well as making the greatest contribution to the community’s safety and sense of security. In addition, the in-vehicle costs are the lowest because the complexity of such systems is largely at the dispatch or control centre. 5.4 Summary Remarks All positioning applications appear to be growing at a healthy rate, with predictions of the annual value of GNSS products and services to be of the order of several tens of billions of (U.S.) dollars within a few years. As new GNSS technology (signals, codes, etc.) is brought online, new augmentation systems developed, and the basic GNSS technology improves (especially to allow it to be embedded within mobile devices such as mobilephones and PDAs), this will in turn give an even greater boost to consumer applications. In fact almost all GNSS market studies have suggested that the most dramatic increases will be in the market for Telematics products and services, such as for personal location and transport telematics. GNSS can be viewed as an enabling technology, that will be the spur for the development of a new generation of products and services for the mobile society. We conclude this paper with the following statement that encapsulates the vision of a future for such a mobile society: "In the same way that no one nowadays can ignore the time of day, in future no one will be able to do without knowledge of their precise location." REFERENCES Catling I (1994) Advanced Technology for Road Transport: IVHS and ATT, Artech House, Boston. Crosby GS, Ely WS, McPherson KW, Stewart JM, Kraus DK, Cashin TP, Bean KW, Elrod BD (2000) A Ground-based Regional Augmentation System (GRAS) – The Australian proposal, 13th Int. Tech. Meeting of the Satellite Division of the U.S. Inst. of Navigation, Salt Lake City, Utah, 19-22 September, 713-721. Djuknic GM, Richton RE (2000) Geolocation and Assisted-GPS, online available HTTP: (accessed 15 September 2002). Drane CR, Rizos C (1998) Positioning Systems in Intelligent Transportation Systems, Artech House, Boston, London. Evans AG, Swift ER, Cunningham JP, Hill RW, Blewitt G, Yunck TP, Lichten SM, Hatch RR, Malys S, Bossler J (2002) The Global Positioning System Geodesy Odyssey, Navigation, 49(1), 7-34. Grejner-Brzezinska D (2001a) Mobile mapping technology: Ten years later (part 1), Surveying & Land Information Systems, 61(2), 79-94. Grejner-Brzezinska D (2001b) Mobile mapping technology: Ten years later (part 2), Surveying & Land Information Systems, 61(3), 137-151. Grejner-Brzezinska D (2002) GPS instrumentation issues, Chapter 10 in Manual of Geospatial Science and Technology, J. Bossler, J. Jenson, R. McMaster & C. Rizos (eds.), Taylor & Francis Inc., 127-145.
Grejner-Brzezinska D (2003) GPS instrumentation issues, Chapter 4 in Telegeoinformatics: Location-Based Computing and Services, H.A. Karimi & A. Hammad (eds.), Taylor & Francis Inc., to be published late-2003. Hein GW, Godet J, Issler JL, Martin JC, Erhard P, Lucas-Rodriguez R, Pratt T (2003) Galileo frequency & signal design, GPS World, 14(6), 30-37. Hjelm J (2002) Creating Location Services for the Wireless Web, John Wiley & Sons Professional Developers Guide Series, New York. Int. Assoc. of Lighthouse Authorities (1999) IALA Guidelines for the Performance and Monitoring of a DGNSS Service in the Band 283.5-325 kHz, March. Karimi HA, Hammad A (eds.) (2003) Telegeoinformatics: Location-Based Computing and Services, Taylor & Francis Inc., to be published late-2003. MIT Technology Review (2003) WhereWare, September, 46-52. Petrovski IG, Ishii M, Torimoto H, Kishimoto H, Furukawa T, Saito M, Tanaka T, Maeda H (2003) QZSS – Japan’s new integrated communication and positioning servuice for mobile users, GPS World, 14(6), 24-29. Rizos C (2002a) Introducing the Global Positioning System, Chapter 7 in Manual of Geospatial Science and Technology, J. Bossler, J. Jenson, R. McMaster & C. Rizos (eds.), Taylor & Francis Inc., 77-94. Rizos C (2002b) Network RTK research and implementation: A geodetic perspective, Journal of Global Positioning Systems, 1(2), 144-150. Rizos C, Drane CR (2003) The role of telegeoinformatics in ITS, Chapter 12 in Telegeoinformatics: Location-Based Computing and Services, H.A. Karimi & A. Hammad (eds.), Taylor & Francis Inc., to be published late-2003. Seeber G (2003) Satellite Geodesy, de Gruyter, 2nd ed., Berlin. Spilker JJ, Van Direndonck AJ (2001) Proposed new L5 civil GPS codes, Navigation, 48(3), 135-143. U.S. Institute of Navigation (2003) Present and future aspects: GLONASS, GPS and Galileo: A multi-expert interview, 13(1), Spring edition quarterly newsletter. Volpe National Transportation Systems Center (2001) Vulnerability Assessment of the Transportation Infrastructure Relying on the Global Positioning System, online available HTTP: (accessed 26 September 2003).
