[time-nuts] New Japanese GPS accuracy
jimlux at earthlink.net
Tue Jan 4 22:34:24 UTC 2011
On 1/4/11 12:53 PM, Hal Murray wrote:
> jimlux at earthlink.net said:
>> Similar in concept to waas or tass, the satellite provides a nav signal and
>> differential corrections.
>> One of the goals is to make a nav system that performs well (sub meter) in
>> urban canyons, which conventional gps does not
> I thought the idea with waas was to correct for the delays through the
> ionosphere by measuring the error at a known (nearby?) location and
> broadcasting the correction. The idea is that a nearby location would have
> similar delays and similar errors.
> I thought the problem with urban canyons was multi-path and blocked signal.
> How is a correction for ionospheric delays going to help that?
> I must be missing something interesting.
> phk at phk.freebsd.dk said:
>>> As far as I know, it is a geo-synchronous polar orbiting D-GPS system.
>> Duh! "Sun-synchronous" of course.
> The original crunchgear article said they needed 3 satellites to get 24 hour
> I can't picture an orbit pattern that's going to use 3 satellites. Geosync
> would work with one satellite, but Japan is fairly far north. Are they doing
> something like picking the orbit height and inclination angle so that the
> satellite period is 24 hours and over Japan rather than the equator at the
> right time?
elliptical orbit (a'la Molniya) so they appear above 70 degrees
elevation 12 hours a day.
Also interesting that they are trying to figure out a way to do
precision timekeeping on orbit without using atomic clocks. Space
flight qualified atomic clocks are pretty obviously a export controlled
kind of thing, not to mention really hard to do in the first place, so
maybe they're trying to avoid having to do the development. It's one of
those things where the basic principles are well known, but I'll bet
there's an awful lot of art in building a Cs clock. Heck, there's a lot
of art in building a spaceflight qualified quartz oscillator. (For
instance, has anyone homebuilt a Cs clock?)
More information about the time-nuts