Woodstramp, on 16 August 2011 - 10:04 AM, said:
Like I told Ecanderson...sounds like even if you have transmission filters that were salavaged from the alien spacecraft they keep at Area 51, the transmitter running high enough power will still be like dropping boulders in a pond that is used to gentle sound of rain drops.
Understanding the various stages of a radio receiver, and that's really what's on the front end of any GPS, will help to understand why even filtering may be insufficient if a unit is close enough to a LS transmitter.
The very first (RF amplifier, AKA "front end") stage of a receiver is designed with a certain amount of gain (amplification) in mind based upon the strength of the desired signal and the signal to noise ratio needed to keep the signal intelligible. The sensitivity of a receiver is in no small part a function of the very first RF stage that grabs the signal from the receiver's antenna and amplifies it. At a distance, any radio signal is surprisingly weak, and it's quite the miracle that receivers can be designed to take those weak signals and do something with them. In the case of a GPS receiver, the signals being received are (relative to other signals) VERY weak, coming from a very low power device half a bazillion miles up there, and with an antenna that has to cover a few degrees of arc and spread its signal around a fair bit of turf, making the result in any one spot on the ground weaker still.
A subsequent stage takes the amplified signal and tries to pick out a particular center frequency from the million different signals that are being received. If the first stage or this stage is being overloaded due to a signal that is far more powerful than the receiver was designed to manage, things can go badly here, even before we reach other stages. Some of the very strong adjacent channel signal will push its way through, and the result will be distorted to boot. In a radio designed to pick up audio, it will just sound lousy. On a GPS receiver, you'll get an unintelligible data signal that can't be used to figure out what satellite is talking, much less compute a position.
Apart from ANY filtering that might or might not be employed downstream, it is possible to simply overload the front end of a receiver to the point where any signal will be distorted beyond use. Using sharper filters really doesn't respond to the problem. Given the spectrum involved, there would always be a problem if you get close enough to a LS facility. The working group showed that under the original plan, "close" wasn't very close at all, and was measured in miles. Attempting to improve on this would have been only to try to reduce by some amount the distance at which the LS equipment was interfering with the GPS signals, not to eliminate the problem.
There's no such animal as an RF filter that has 90 degree slope (square) sides - the perfect bandpass filter. There's always some slope on the sides. To avoid attenuating desirable signals on frequencies that are toward the edge of the band, a certain amount of signal from outside of the band is allowed in as well. The problem occurs when the undesirable signal is orders of magnitude stronger than the signal you want to keep. THAT WAS NEVER SUPPOSED TO HAPPEN IN THIS CASE. A little safe spacing (a 'guardband') had been established long ago in the frequency allocations, and everyone designed equipment accordingly. When the FCC decided to let LS deviate from previously approved use of that bit of spectrum in the way that LS asked, the FCC could
have known the results would not be pretty. They knew full well the kind of gear that was being used out here in the real world, and ought to have anticipated the results. Either they're technically incompetent and did not anticipate this, or they're a bunch of goofs who sold out an entire industry to promote an agenda. So the leadership of the FCC either winds up looking like a bunch of idiots, or a bunch of idealogues. Take your pick. Neither is acceptable.
This post has been edited by ecanderson: 16 August 2011 - 02:54 PM