I managed to get my Blackberry into "engineering" mode, which I think
really is hacker mode. Here is what it dumped for my GPS (one of many
screens). OK, exact location altered to protect the innocent. Feel free
to espouse on the meaning of the parameters.

Are the frequencies and table look ups some sort of temperature
compensation for an oscillator?

GPS API requests:
None
GPS fix requirement:
None
GPS state:
Off
GPS Last Fix:
Latitude: xx.834961 N
Longitude: xxx.489552 W
Altitude: +259 m
Velocity: 0 km/h
Direction: N (14 deg)
Uncertainty: 2 m
UTC Time: 2011/10/08 20:19:57
Satellites: 8
CNo: 33 32 30 28 26 25 24 23
SVs: 12 4 17 25 5 10 9 2
Age: 100 mins
TTFF Information:
TTFF: 28 s
Successes: 1 of 1 tries
Average TTFF: 28 s
Satellite Tracking:
Average CNo: 0
....
....
....
....
....
....
....
....
....
....
....
....
Aiding pushed:
Location (local) (102 mins ago)
Time (local) (102 mins ago)
Ephemeris (local) (102 mins ago)
Ionospheric (local) (102 mins ago)
Local time aiding:
State: Valid
System time error: 0 s
Local ephemeris aiding:
Last EE request: 102 mins ago
Last EE delivery: 0 s from request
Frequency Aiding:
Last check:
Actual: 661 Hz
Table: 574 ± 787 Hz (75)
Error: 87 Hz (11% of margin)
Worst check:
Actual: 665 Hz
Table: 565 ± 787 Hz (75)
Error: 100 Hz (12% of margin)
SD Card Logging: (Disabled)
Power Management Info:
Time in APM mode: 0%

On Oct 8, 11:23*pm, miso <m...@sushi.com> wrote:
>
> Are the frequencies and table look ups some sort of temperature
> compensation for an oscillator?
>

Short answer: seems likely

Longer answer:

Acquiring a GPS signal inolves a search in both time and frequency.
Data Fusion Corp. posts a nice description of a generic receiver here,
with signal acquisition beginning on pg. 4:

http://datafusion.com/gps/baselinereceiver.pdf

"The search space must cover the full range of uncertainty in the code
and Doppler offset. . . . The range of the Doppler dimension is
governed by the vehicle and GPS satellite dynamics and the stability
of the receiver oscillator. For a terrestrial user system this is
typically in the range of 5 to 10 kHz."

For an assisted GPS (A-GPS) system, which is one mode of Blackberry
GPS systems, the base station can send approximate satellite Doppler
data to the mobile. Thus the only frequency uncertainty is from the
mobile's oscillator. However, a 1 ppm oscillator will have a
frequency uncertainty of 1.6 kHz at L1, which would still require
searching several frequency bins, each with frequency widths of a few
hundred Hz.

After an autonomous (unaided) GPS signal acquisition, the receiver can
calibrate its oscillator frequency from delta range measurements and
its oscillator phase (time) from pseudorange measurements. It may
save these calibration values to try to model oscillator drift, and
apply that model on subsequent acquisitions.

Tom Clark explains this more directly here:

http://gpsinformation.net/main/gpslock.htm

"The usual search strategy is to start at an oscillator frequency
predicted on the basis of
(a) what is my approximate position? (based on either the last known
position or the guess you "typed in"),
(b) where are the satellites now and what is their apparent velocity
(based on the almanac data downlinked from the satellite and stored in
the receiver's memory), and
(c) what is the best guess for the receiver's crystal oscillator error
(based on the last known error and the current temperature). "

Some A-GPS systems have the base station send a single tone to the
mobile to let it calibrate its oscillator. This is described in a
1997 patent by Snaptrack (now part of Qualcomm):

http://www.freepatentsonline.com/5781156.html

http://www.freepatentsonline.com/pdf.../US5781156.pdf