SA_AS_243

HOW GOOD IS GPS?

The Civilian - Military Relationship and the Impact on GPS Performance

Although GPS is a military navigation system, the civilian sector represents an important (and rapidly growing) user group that has increasingly lobbied the U.S. Government in order to influence: (a) the direction of GPS system development; (b) official GPS policy concerning enhancement and control; and (c) the follow-on systems to GPS for the 21st century.

Several policy decisions have already been made which impact on GPS performance. Some of these actions were agreed to during the early system design phase, while others were invoked only after much of the present system had been deployed:

Two PRN ranging codes are implemented. The C/A code is intended for general, civilian use, while the P-code was reserved for U.S. military and other authorised users. The important distinction between the codes was that as a result of the P code's tenfold higher measurement resolution, it was expected that the accuracy of positioning using the P code would be better than that possible using the C/A code by a similar margin. However, to everyone's surprise, the performance of C/A code positioning was often no worse than by a factor of two compared to P code positioning. (The latest C/A code tracking technology promises ranging quality almost as good as P code ranging.)
Two levels of positioning performance were therefore designed into the GPS system from the very beginning. The positioning service based on using C/A code ranging data is the Standard Positioning Service (SPS), while the service based on P code ranging data is the Precise Positioning Service (PPS).
As a result of the surprisingly good level of SPS accuracy, the policy of Selective Availability (SA) was endorsed in order to artificially widen the gap between the two positioning services. SA is an intentional degradation of the accuracy of GPS horizontal positioning to 100m (at the 95% confidence level), and height determination to 150-170m (at the 95% confidence level), for SPS users (see Figure below). SA has been implemented since 25 March, 1990, through: (a) a corruption of the transmitted ephemeris data within the Navigation Message (the so-called "epsilon" effect), and (b) the satellite clock is "dithered" (the so-called "delta" effect). SA does not affect PPS users who have GPS equipment able to decipher the correct ephemeris and clock error data.
Prior to 1994 any GPS hardware manufacturer was able to develop a P code ranging receiver (the P code PRN generation algorithm having been published). The non-military market for P code receivers was always assumed to be very small, and one that could be controlled by the U.S. Government through the issuance of "export licences", etc. However, a significant demand by the surveying market for dual-frequency phase tracking GPS receivers led to an expansion in the production of P code capable positioning equipment.
Under another policy known as Anti-Spoofing (AS), access is denied to the P code modulated on both L-band frequencies. AS was implemented on 31 January, 1994, through the encryption of a further secret "W code" onto the P code. The rationale behind this decision was that by keeping the military PRN code secret, an enemy of the U.S. could not jam the signal using a ground-based transmitter, nor "spoof" military GPS receivers by transmitting a false P code signal from a satellite. However, several GPS receiver manufacturers have developed proprietary techniques for making dual-frequency measurements even in the presence of AS (section 4.2.3).
Dual-frequency observations will lead to more accurate positioning results than single frequency observations because the ionospheric bias can be eliminated from the code range measurements. The fact that the C/A code is only modulated on the L1 carrier is therefore an intentional design decision to ensure that the SPS service cannot approach the quality of the PPS service, even with SA off.

SELECTIVE AVAILABILITY:

Policy of intentional degradation of SPS navigation accuracy introduced on 25 March, 1990.

100m horizontal accuracy (95% confidence level), 150-170m height accuracy.

Horizontal accuracy better than 300m 99.99% of the time.

60m Spherical Error Probable.

0.3 m/sec velocity accuracy and approximately 340nsec time transfer accuracy.

Classified algorithm and characteristics (periodicities, etc.).

Military receivers are able to overcome SA.

Two effects:

(a) "dithering" of the satellite clock (the "" component), and (b) errors in the broadcast ephemerides (the "" component).

Can be overcome by using differential GPS positioning mode.

It must be emphasised that the situation concerning the SA and AS policies are under almost continuous review. The "tug-a-war" situation between the GPS system controllers and civilian users needs to be resolved, and the reader is referred to LACHAPELLE (1995) and SANDLIN et al (1995) regarding the outcome of studies into alternative models of joint civilian-military GPS operation and recommendations concerning SA and AS. In particular, it has been recommended that SA be deactivated within the next few years, and that another frequency be transmitted to support civilian dual-frequency operations.

SA also biases carrier phase measurements (section 6.2.4). However, the policy of SA is explicitly aimed at the real-time point positioning (that is, the navigation) user, and has only a marginal effect on those users that have adopted differential GPS techniques, in either the real-time (precise navigation) or the post-processed mode (in general, surveying), based on pseudo-range and/or carrier phase data.

ANTI-SPOOFING:

Satellite signal design intended to prevent military receivers from being "spoofed" (by tracking "false" GPS satellites).

P code is replaced by a secret Y code.

P code receivers will not function when AS is on.

Only receivers with the appropriate "key" can use the Y code for ranging.

Implemented on 31 January, 1994.