Despite the advantages of an airborne measurement platform, the widespread use of aerogeophysics and the development of new techniques has been hampered by the inability to accurately position an aircraft. Traditionally it has been difficult to accurately position an aircraft in three dimensions to better than 100 m in the horizontal and 200 m vertically. Poor positioning simply made many measurements impossible. Accurate measurements of the land surface below an aircraft and the measurement of the gravity field were very difficult, if not impossible [ Nettleton et al., 1960]. High resolution surveys which require closely spaced flight lines have also been difficult due to the inherent ambiguity in aircraft positions.
In the last decade, the advent of the Global Positioning System (GPS) has broadened the applications of airborne geophysics due to the enhanced ability to both navigate and position an aircraft. The refinement of differential positioning techniques using the GPS system in the last four years has further expanded the use of airborne geophysical applications.
GPS is system of satellites which can be used to position an aircraft to sub-meter accuracy at any point on the globe when used in the differential mode. GPS is a powerful, evolving tool for both aircraft positioning, navigation and potentially, orientation recovery. The system has the advantage of being globally accessible, functioning independent of the local weather conditions and having relatively low user costs. The GPS system satellites transmit bi-phase encoded signals at two frequencies denoted by L1 (1.575 GHz, 1GHz = 109 Hz) and L2 (1.227 GHz). Two basic codes are written on the GPS carrier signals at two frequencies: the course acquisition code (C/A code) and the precise positioning code (P-code). The system was designed to provide navigation to an accuracy of approximately 10 meters with the P-code and approximately 100 m with the C/A code.
Researchers accessing the full range of signals transmitted by the GPS satellites have expanded the capabilities for positioning moving vehicles, such as aircraft, to the sub-meter level. These applications have extended beyond the initial design of the system which was based principally on the decryption of the two codes. The three fundamental GPS observables used to precisely position research aircraft are: the pseudorange measurements, the carrier phase and the Doppler shift.
The pseudorange measurements are the measured difference between the arrival time of a GPS signal (as measured on the receiver clock) and its transmission time (as determined by the satellite clock). By simultaneously observing four or more satellites, it is possible to determine the position of the receiver and correct for differences in time between the receiver's clock and the GPS satellite time system. Selective Availability (SA) and the inability for civilian users to access the P-code psuedorange measurements (Anti-spoofing, AS) reduce the accuracy of GPS psuedorange positions. The ephemeris error and clock dithering introduced by SA reduces the psuedorange position accuracies to 60 m. Carrier phase measurements are made by reconstructing the carrier signal. This process includes removing the bi-phase encoding and measuring the phase difference between the reconstructed carrier and a local oscillator within the receiver for both frequencies. As this carrier-beat phase rotates through cycles, the number of cycles is accumulated. The phase measurement is the accumulated phase change from the time the satellite is acquired by a receiver (locked on) until the receiver loses the signal (loss of lock). The carrier phase measurement has a noise level of a few millimeters. The critical parameters for carrier phase measurements are the resolution of the ambiguity and the necessity of maintaining continuous lock on the satellite signal. Ambiguity resolution is the determination of the initial number of integer cycles between the receiver and the satellite in number of cycles [i.e. Lachappelle et al., 1991]. Loss of lock on the satellite signal causes an uncertainty to the measurement called a cycle slip. Resolution of the uncertainty introduced by cycle slips has been a major error source in carrier phase measurements. Improvement in receiver tracking algorithms has reduced the number of cycle slips apparent in a typical airborne survey.
The Doppler shift is most often determined from the time derivative of the carrier phase. The Doppler observable, which is rarely used for geodetic positioning, can be useful for connecting phase measurements across small gaps or cycle slips (<5s) when the signal from a satellite is lost (e.g. during banked turns or rapid accelerations).
Together these three observables provide the fundamental suite of tools for resolving accurately the position of an aircraft. The principal errors in a GPS position arise from errors in timing and errors introduced due to the propagation of the signal from the satellite to the receiver. SA (Selective Availability), the high frequency dithering of the satellite clock by up to 0.2 microseconds and corruption of the broadcast navigation message is a clear example of an induced timing error. The dithering cannot be removed without access to a classified decryption key. The SA error size is much larger than the timing errors which are intrinsic to the GPS clocks. The propagation errors are associated with dispersive delay in the ionosphere, atmospheric delays and the signal bouncing off objects surrounding the antennae before it is recorded by the receiver (multipath). Propagation errors are amplified in high latitude applications due to enhanced ionospheric activity and the low azimuth of the GPS satellites. Differential techniques are widely used to minimize many of these errors.
Differential techniques are based on the concept that many of the errors in timing and propagation can be eliminated by observing each satellite at a second receiver in a fixed location [ Wells et al., 1987; Hofmann-Wellenhof et al., 1993; Seeber, 1993]. The approach assumes that systematic errors seen by both receivers, such as the dispersive delay in the ionosphere, can be removed by differencing the signal. Some error sources which are not common to both receivers such as multipath, cannot be removed with this technique. This approach can be applied to both psuedorange and carrier phase measurements. The accuracy of differential positions decreases with increasing baseline length, the distance between the roving aircraft receiver and the fixed receiver. Differential psuedorange positions are accurate to 0.5-3.0 m depending upon the receiver type used. This technique is being used experimentally for commercial aircraft approach systems. The implementation of AS restricts this technique to CA code measurements. Differential carrier phase measurements are as accurate as 10-20 cm and require extensive post-mission analysis [ Mader and Lucas, 1989; Mader, 1992]. Differential carrier phase positions are unaffected by SA errors if the satellite ephemeris is determined independently of the broadcast signal although the rapid fluctuation in the satellite clocks complicates the correction of cycle slips. The impact of AS on differential carrier phase measurements is that a longer averaging time, up to 1 second, is required and that codeless tracking algorithms introduce a greater number of cycle slips. Real-time differential systems using carrier phase measurements are under development [ Frodge et al., 1994].
Several efforts have been made to demonstrate the robustness of the differential GPS approach for accurately locating aircraft. The most difficult component to constrain is the vertical position. In experiments over water and ice the vertical position of the aircraft recovered from independent systems is within 20-50 cm of the elevation measurements recovered from the differential GPS carrier phase measurements [ Brozena et al., 1989; Blankenship et al., 1993]. Three airborne geophysical measurements have evolved rapidly with the advent of precise GPS positioning including; airborne gravity, airborne topographic measurements and high resolution surveys.