On Jan 20, 12:08*am, "Ed M." wrote:
There's a lot of literature describing how a precise clock can
mitigate the strong correlation between the clock and vertical
position states, hence improve VDOP, improve integrity monitoring,
etc. *Pratap Misra has written quite a bit on the topic.

An early paper on clock coasting (there were earlier experiments):

Sturza, Mark A., "GPS Navigation Using Tbree Satellites and a Precise
Clock", NAVIGATION, Vol. 30, No. 2, Summer 1983, pp. 146-156,http://www.3csysco.com/Pubs/GPS%20Na...Three%20Satell...

Can't find an on-line copy of this one:

“The Role of the Clock in a GPS Receiver” by P.N. Misra in GPS World,
Vol. 7, No. 4, April 1996, pp. 60–66.

A dissertation that involved flight testing with a Boeing 767 at the
FAA's Atlantic City test site:

Kline, Paul A., "Atomic Clock Augmentation For Receivers Using the
Global Positioning System," Ph.D. Dissertation, Virginia Polytechnic
University, 1997,http://scholar.lib.vt.edu/theses/ava...2516142975720/

A thesis sponsored by Misra:

Sean G. Bednarz, "Adaptive Modeling of GPS Receiver Clock for
Integrity Monitoring During Precision Approaches," M.S. Thesis,
Massachusetts Institute of Technology, 2004,http://dspace.mit.edu/handle/1721.1/17756

A GPS World Innovations column from 2007:

http://tf.boulder.nist.gov/general/pdf/2267.pdf

Coasting studies I've read seem to assume usage of a worse CSAC than a
Cs atomic clock, something with at least one order of magnitude less
stability than your typical Cs atomic clock. Taking Symmetricon's
claims at face value here... Top of the line quartz oscillators = 30ms
per year stability, four orders of magnitude better would be roughly 3
micro seconds per year stability, 8 ns per day stability, or 1 ns
every 4 hours. Or 10 cm per hour drift. In an unjammed environment,
today even in the middle of a GPS maintenance event, PDOP spikes don't
last 2 hrs for the same area. And a 20cm error would be acceptable
even for a CAT I autoland (VPL around 10 meters). Expecting the need
to coast for more than 2 hours would be an nuclear war or extreme
military jam requirement, not a civilian need.

The most exciting aspect is this is only the first generation
implementation. There are no competitors yet. 2nd generation should be
here in lets guess 5 yrs. Imagine a CSAC on par with current Rb atomic
clock for short term stability ! This not only helps with GNSS
applications, but also helps every radio transmitter/receiver, 4g/
WiMax cell towers, ultra high speed laser transceivers, the list goes
on and on. CSACs will directly replace current 10 MHz GNSS based
frequency standards, used on any serious data communication/telecom
environment, even small rural ISPs are more and more using WiMax or
high end WiFi towers that require a more accurate frequency standard
than a quartz oscillator. 5g cell systems might use upcoming 100 MHz
atomic clock frequency standard to improve current RF performance by a
mile.

Marcelo Pacheco