11.4.5  Time Transfer Using GPS

Time transfers (clock comparisons) may be made in a number of ways using the GPS

satellites.  The time dissemination process described in paragraph 11.4.4 is a "passive"

method, in which the user acquires an accurate time reference without having to

transmit timing signals or data.  Other ways that can provide more accu rate

comparisons are described in this section.

11.4.5.1  Coordinated Simultaneous

 View Time Transfer

In this method, a pair of stations simultaneously observes the same satellite(s); then

(through some communications medium) they exchange readings of their local clock

time against the time disseminated by GPS.  The difference between these readings is

quite accurately the difference between the stations  clocks.  The satellite clock is

primarily a transfer clock and does not directly affect the time transfer accuracy.  This

method might be used where the user clocks are required to maintain time or frequency

agreement more precisely than UTC can be disseminated through GPS.  The method

works particularly well when the participating clocks are located reasonably close

together (within some hundreds of kilometers).  The method can also substan tially

reduce the effects of S/A on time transfers made with the C/A code, because both

ephemeris and ionospheric effects are reduced. Unless the time transfer is made with

USNO or a UTC(USNO) traceable reference, the result is relative rather than absolute

time accuracy.

11.4.5.2  Coordinated Simultaneous

 View Time Transfer with USNO

USNO uses a coordinated simultaneous  view method as shown in Figure 11 5, to

provide more accurate UTC(USNO) to certain Pre cise Time Stations within

simultaneous view range.  Both USNO and the distant observer track the same GPS

satellite(s), derive UTC (USNO) from the satellite s NAV msg and pseudorange

measurements, and compare this time with the time maintained by t heir local atomic

clocks.  USNO compares UTC (USNO) derived from the GPS satellites with the USNO

Master Clock.  Thus, USNO can deter mine the Control Segment and Satellite   induced

errors that the observer will have in his GPS  derived UTC(USNO).  The distant

observer can then correct his GPS derived UTC (USNO) with correc tions received from

USNO via a data link.  Now the distant observer can correct his clock very accurately to

serve as a local reference traceable to UTC (USNO). The time accuracies that can be

obtained by this method are shown in Table 11 1.  The Table 11  1 values are valid for

time transfer using C/A code only when SA is switched off. Smoothing of the time

measurements brings the error down to what can be expected for a P code receiver.

The errors due to ephemeris uncertainties and ionospheric delay usually cancel out of

if the two receivers are close to each other.  This is because they have nearly the same

line of sight to the satellites, and the signals travel through the same part of the iono  

sphere.  In some cases where the two receivers are close to each other, the use of both

L1 and L2 to compensate for iono spheric delay will be less accurate than not correcting

for ionospheric delay at all.  This is due to the fact that dual frequency com pensation

for ionospheric delay is not perfect, and use of the ionospheric delay broad cast by the

satellite by both parties produces more accurate results.  For most cases where

11 14