Prof. Chris Rizos
School of Surveying and Spatial Information Systems,
University of New South Wales, Sydney, Australia
Email: [email protected]
The U.S. Global Positioning System (GPS) and the USSR (now Russian) system Glonass have operated for two decades. Europe is developing Galileo and China launched Beidou – a Regional Navigation Satellite System (RNSS) – and then announced Compass, a Global Navigation Satellite System (GNSS). Japan will soon launch the first satellite of its own regional augmentation to GPS and Galileo known as the Quasi-Zenith Satellite System (QZSS). India has proposed the Indian RNSS (IRNSS). In addition there are a number of Space-Based Augmentation Systems (SBAS) that are already deployed, or will be soon, that broadcast extra navigation signals primarily intended for aviation users. Most of the big space players all now have, or will soon launch, a Navigation Satellite System (NSS) (which in this paper is the generic acronym for GNSS, RNSS and SBAS).
The main advantage that new NSSs bring is that they provide more satellites. It is estimated that by 2013-15, there will be up to three times the number of satellites and four to six times the number of individual sig- nals on which measurements can be made, compared to today (approx 50 satellites, on which two frequency measurements can be made with appropriate hardware). It is generally conceded that more satellites and signals there are, the better the positioning performance is (in terms of accuracy, availability and integrity). Indeed, the advantage that the ‘extra’ Glonass satellites provide for GNSSRTK has seen all high-end manufacturers offer GPS+Glonass receivers, and increasingly claim their products are also Galileo-ready. Hence this rapid increase in satellite signals and constellations over the next five years means that a receiver that can exploit all of the new signals may be the ultimate in satellite positioning and navigation – a so-called ‘system of systems’ (SoS) receiver. Figure 1 shows the high average (24hr) visibility in the Asia-Oceania region of the future NSSs.
However, it must be emphasised that there will be many user communities, perhaps a vast majority that will only use a subset of the available signals and constellations. In particular, there may be regional markets/products that will take advantage of signals broadcast only over their area – and they may even be mandated to do so by the governments of the NSS ‘signal providers’. This raises many interesting questions, as the world has only, to date – with the exception of the few Beidou terminals in China – been using ‘global’ GNSS products. How the various RNSS and SBAS signals will be incorporated into user products and services is still unknown.
While there are hardware design issues to be addressed, the more challenging are the market-specific issues such as applications software, the trend to regionalisation of navigation systems, and business models, marketing and service support. The goal of the International Committee on GNSS (ICG – https://www.unoosa.org/oosa/en/ SAP/gnss/icg.html) is, however, to ensure sufficient interoperability of NSSs that SoS receiver products – or products using any subset combination of NSS signals – can be relatively easily developed, and do not simply consist of a number of separate NSS receivers inside a single box.
Market forces can be expected to ultimately define the receiver configuration for most consumer devices such as car navigation systems (transport telematics applications), PNDs, and mobile phones (LBS applications). However, for the high-end applications (synonymous in this paper with “high accuracy” – intended to address surveying, machine guidance and geodesy applications) that use differential GNSS techniques, the situation is a little more complex. Such applications require appropriately configured permanent reference station networks. Hence consideration of the impact of multi-constellation NSS is not just a matter of satellites and user terminals, but must also include the terrestrial positioning infrastructure.
Implications of Extra NSS Satellites and Signals
By 2013-15, appropriately equipped users would benefit from:
- enhanced accuracy (more observations, greater measurement redundancy, faster solution filter convergence, lower PDOP, etc.),
- improved availability (about three times more visible satellites, dual- and triple-frequency signal availability, and for high-end users more rapid ambiguity resolution and lower constraints regarding user-reference receiver separations), and
- higher integrity (high measurement redundancy, lower interference vulnerability, enhanced QC algorithms, etc.).
All high accuracy GNSS techniques currently rely on the differential mode of operation, in which carrier phase measurements from multiple frequencies are processed (for both carrier ambiguity resolution and positioning computations), hence there are two aspects to multi-constellation NSS. One is the receiver itself, either operating as a user receiver or as a reference station, and the other is the design of the continuously operating reference station (CORS) positioning infrastructure. For example, at first glance, the CORS infrastructure should be designed to support differential positioning by operating reference receivers with at least the same level of signal tracking capability as those of its users. However, the dilemma is:
- SoS receivers will be the most expensive hardware on the market. Network savings could be made by having a less dense CORS network, however, only users operating similar SoS receivers could take advantage of such a sparse CORS infrastructure. Users with simpler (lower cost) hardware, such as dual-frequency receivers, would be disadvantaged as the nearest reference station could be too far away to ensure reliable and rapid ambiguity resolution.
- The lower cost dual-frequency receivers could be deployed as reference receivers with spacings of tens of kilometres, similar to today’s CORS supporting Network-RTK and GNSS-RTK to support users seeking cm-level relative accuracy. However, such a CORS network could not service the users operating high-end SoS receivers.
Future “System-of-Systems” Receiver Design
The following comments can be made with regard to future SoS receivers:
- A multi-frequency, multi-constellation SoS receiver would be expensive, especially if it were to only address the niche markets for high-end users.
- The power consumption of such SoS receivers may be so high that they would be unlikely to be used for portable applications relying on battery power alone, possibly restricting its use to heavy equipment to support machine guidance in agriculture, mining and construction, and for scientific uses.
- Such a receiver would be expected to be the “receiver-ofchoice” as a reference station receiver, so as to service the needs of all users, including those operating full SoS receivers.
- Increasingly such full SoS receivers may perform all their baseband processing using software-based correlators, which will to some extent future-proof the receivers as GNSS satellites with new signals are launched in rapid succession.
- There will be many tradeoffs made in user receiver equipment design, between the ultimate performance of a full SoS receiver and a receiver that may track, for example, just two interoperable frequencies such as L1 and L5, resulting in user equipment that tracks some of the signals of most/all the satellites.
Future CORS Infrastructure Design
It is important to recognise the significant contribution of the “super-network” of reference stations of the International GNSS Service (IGS) to geodesy, and to the GNSS community in general. Several hundred globally distributed CORS (increasingly with Glonass tracking capability) operate on a continuous basis, many for over ten years, contributing data to the IGS analysis centres (https://igs.org) and other users. The IGS was established in January 1994 as a service of the International Association of Geodesy. Since June 1992 the IGS originally known as the International GPS Service for Geodynamics, from 1999 simply as the International GPS Service, and finally since March 2005 as the International GNSS Service – has been making freely available to all users: (a) raw GNSS tracking data from its global CORS network, and (b) high accuracy satellite ephemerides and other derived products.
The IGS activities are fundamental to scientific disciplines concerned with climate, surface weather, sea level change, gravity, space weather research, and more. The IGS CORS network therefore pro- vides the fundamental NSS positioning infrastructure for science and society. This is densified at the regional, national and local level with many more CORS receivers at spacings ranging from a few tens of kms (to support standard GNSS-RTK) to hundreds of kms (for scientific applications). Unfortunately the Asian region is amongst the sparsest regions in the world as far as the IGS network is concerned (see Fig 2).
At present, there is a strong trend to the establishment of more and more CORS networks to support geodesy, surveying and precise positioning applications in general, although it is not clear who will actually own, operate or manage this precise positioning infrastructure in the future. It is worth speculating on what a future CORS infrastructure may consist of. One scenario could be:
- Full SoS CORS receivers able to track all NSS signals, established with relatively large inter-receiver separations, perhaps of the order of several hundred kms or more. These may be the future backbone of the IGS, as well as being the receiver of choice for the fundamental national geospatial reference frame stations. Plus…
- Lower cost multi-constellation NSS CORS receivers to densify the fundamental CORS network, probably with dual-frequency tracking capability established at closer receiver spacing, from just several kms apart (to support structural deformation monitoring, and single-base RTK applications), up to several tens of kms (to support most differential and Network-RTK users, and possibly atmospheric remote sensing applications). The interoperable frequencies are most likely to be the L1 and L5 frequencies which all NSS are likely to be transmitting.
Such a mixed CORS network design could service all users requiring high accuracy positioning, no matter what the tracking capability of their receiver is. It should be emphasised however that most of the new NSS signals will not be transmitted before 2013 at the earliest, hence the current investment in CORS infrastructure for the IGS will continue to be in GPS+Glonass capable receivers. However, the upgrade of the CORS infrastructure at, or after, 2014 will have to incorporate multi-constellation NSS tracking capability. The market for high accuracy user receivers, on the other hand, will be broad, and several basic multi-frequency configurations are likely – from full SoS receivers for the most mission-critical applications to receivers tracking a subset of all possible NSS signals and frequencies. It is necessary to research the capabilities of the GNSS/RNSS/SBAS constellations now, including in-field testing of multi-constellation user receivers and CORS infrastructure.