Sonoman

Hello NG:
am currently in the transition between the Engineering Design I and
Engineering Design II classes. I am designing a device that needs to get
readings from a GPS sensor. I understand that there is a random error
introduced to the GPS readings by US government regulations. I wanted to
find out if I get two GPS sensors, and I use one static (will be fixed
on a pre-set location) and one that moves around scanning a surface
area, would I be able to find the random error by comparing the two
readings in order to get a more accurate result from the mobile unit
with respect to the static unit? Can I safely assume that if the
readings from the static sensor and the mobile sensor are gathered at
the exact same time they will pick up the very same random error on both
sensors? If not, can they be synchronized to do so?

I do not really need to get an accurate longitude and latitude readings,
but instead I need to get accurate readings with respect to the position
between the two units. Is there a better way to achieve the same result?
Please advise.

Sonoman

Peter

Sonoman wrote:
> Hello NG:
> am currently in the transition between the Engineering Design I and
> Engineering Design II classes. I am designing a device that needs to get
> readings from a GPS sensor. I understand that there is a random error
> introduced to the GPS readings by US government regulations.

There used to be, but your information is a bit out-of-date. Selective
Availability (SA) dithering was turned off in May of 2000.

I wanted to
> find out if I get two GPS sensors, and I use one static (will be fixed
> on a pre-set location) and one that moves around scanning a surface
> area, would I be able to find the random error by comparing the two
> readings in order to get a more accurate result from the mobile unit
> with respect to the static unit? Can I safely assume that if the
> readings from the static sensor and the mobile sensor are gathered at
> the exact same time they will pick up the very same random error on both
> sensors?

No. One problem is that even if the two devices are near each other
they may be using a different set of satellites to determine the
position (just someone walking by or a tree branch may block a
particular sat. from one unit but not the other).

> If not, can they be synchronized to do so?

Read up a bit on Differential GPS (DGPS), which is the proper way to do
this. A reference station at a known location determines the observed
timing error in the signal from each satellite and broadcasts this
information. The mobile unit then uses this information to correct the
satellite data it receives and get a better position determination.
Although SA is no longer a problem there are still errors due to
atmospheric effects, clock errors, ephemeris errors, etc. and DGPS can
help to minimize these.
If you're in North America there is a regional form of DGPS called WAAS
(Wide Area Augmentation System) that uses reference stations throughout
a broad area to model the errors and the resulting data is broadcast by
geosynchronous satellites on the GPS frequency to your receiver. Most
recent GPS models support WAAS (European equivalent is EGNOS, but it
isn't fully operational yet).
>
> I do not really need to get an accurate longitude and latitude readings,
> but instead I need to get accurate readings with respect to the position
> between the two units. Is there a better way to achieve the same result?
> Please advise.

Hard to say without knowing what accuracy you actually need. RTK (Real
Time Kinetic) is another technique used to get much better accuracy than
regular DGPS, but the equipment is rather expensive.