Will, this is an interesting discussion and as Jackdaw says, the devil is in the details. Your idea of using electronically assisted celestial navigation is not new. Satellites, Space Shuttles and even long range missiles have used star trackers for years. I worked for a company that developed them (search CALTRAC Star Tracker). You need a decent view of the sky and a good timing reference and then the proper celestial tables. The trouble with doing that on earth is a lot of stuff gets in the way. Pitching and rolling in a seaway would add challenges but the real problem would be getting a clear view of the sky 24/7. You're making the assumption you can use sensors to detect the very faint radiation emanating from distant stars through overcast or in daylight. The problem with that is mostly to do with antenna gain. The interactions of the RF spectrum through the earth's atmosphere changes quite a bit over the range from DC to light as do the characteristics of the receiving antennas. Star Trackers use radiation near to the visible spectrum. The receiving sensors for these are reasonably small with fairly high gain. When you extrapolate that down to the part of the spectrum that can make it through cloud cover the frequency drops and the wavelength gets longer but the amplitude is quite small meaning you need massive antennas to pick it up. You also need very high gain to pinpoint the source which also means large. As mentioned above, this kind of kit needs to be VERY stable, bolted to bedrock. Even then, trying to discern the radiation from one star to the next would be a challenge and would take some time, scanning some area of the sky with your high gain (narrow field of view) antenna.Jackdaw, I made a bold statement, you called me out on it. I explained myself, you and others pointed to errors in my knowledge-base, I argued back, corrected an assumption based on your information and elaborated on my idea. I remain sceptical about the impossibilities of my thoughts, but I respect your position. If we have the technology to measure a 5mm change in a distant planet's orbit, I can't imagine we can't put instruments on a boat that can find its position within feet of a dangerous and known rock without satellite help.
There is, of course, a point at which this discussion needs to wind down and I think you are wise to be sensitive to it at this time. Another of my many failing is my inability to recognise those times before it is too late.
I hope you know I hold your opinions in high regard. I just don't agree with you on this.
- Will (Dragonfly)
GPS was designed to get around these limitations by providing a known pseudo-random code that the receiver can pick out of the noise. This means you can use a cheap, small, portable, low gain antenna to pick up the signal. In addition, there are some GPS satellites which transmit WAAS which includes ephimeric and ionospheric delay corrections to improve accuracy which stars will never be able to do. The GPS satellites even provides the timing reference for you!
In essence, it is possible to do what you are proposing (in fact it is already being done) but not with the reliability and accuracy of GPS for the given cost and portability using technology available today.