QUESTION:
I realize that a synchronization signal (DS1) derived from a GPS receiver, such as a PRR-10, and one derived from a Cesium PRS, such as a TimeCesium 4500, should be at least Stratum 1 accurate (+/-1X10-11). My question is: Which one would be the more accurate signal and what would be the expected accuracy of each?
ANSWER:
Your question is very well taken. The general "rule of thumb" is that PRS devices based on GPS can provide an accuracy of about 1x10-13. However, to get to this floor (most of them have a "jitter" of about 50 ns) requires an observation interval of about 100,000 seconds (a day). A typical cesium PRS will have an accuracy of about 1x10-12. Generally speaking GPS is "better" in the long-term and cesium is "better" in the short-term. Therefore, if you do a measurement comparing the two, you will be measuring the "error" of GPS in the short-term and the "error" (sometimes called "tilt" because of the way it appears in an MTIE plot) of the cesium in the long-term. Note that cesium devices may vary somewhat from unit to unit and some of them may actually be accurate to 1x10-13 or better (luck of the draw).
|
