Olivier HOUOT wrote: > But isn't that wavelength-dependant antenna size concept a=20 > little artificial? > > There may be some engineering situations where it makes sense > to tie the antenna size to the wavelength, but in the case of > a space probe, i think you try to use the bigger possible > antenna within the constraints of the project (space > available in the rocket nose, weight, probe inertia versus > attitude thrusters power, etc...). Well, sure. But in order to write an equation that is univerally useful, you need to establish certain unchanging criteria. And one of these would be antenna gain. Assume it to be unity in all circumstances, and then use antenna gain to adjust your design. > Speaking of the Shannon's capacity theorem, it is not > dependant on frequency. So what is the hypothesis when saying > SNR will be influenced by this parameter ? If i consider the > same bandwidth and just transpose it to a higher center > frequency then, unless the ambiant noise (natural, > amplifier-generated,...) is very different the resulting > capacity should not change. Correct. Shannon's equation is very simple: Capacity (b/s) =3D bandwidth (Hz) * log2(1 + S(watts)/N(watts)) The center frequency of the channel is not a factor, except as it might impact the SNR or the bandwidth. For example, it's going to me mighty difficult to create a 6 MHz channel if the center frequency is, say, 1500 KHz. For space comms, you also want to get through the ionosphere without too much attenuation, which means you want to stay well above 30 MHz. So these factors are all in play, even if they don't appear directly in the Shannon equation. > On the other hand, higher power can increase the SNR, and > perhaps allow for a less robust, but more efficient > modulation scheme. Are they toying with 8PSK or 16QAM on a > Mars-Earth link? Je n'sais. I think that if 8-PSK has yet to find its way in DBS comms, it's doubtful anyone would expect that much of a power excess to exist in a 6 MHz channel from Mars. But I don't know. > Quantum entanglement (or the EPR paradox) has sometimes > raised hopes for instantaneous communication because, if you > produce two correlated photons and send them in opposite > directions and, at some point, perform a measurement on one > of them, the state of the other one is instantly affected, > whatever the distance between the two. > > Scientists keep saying that you cannot use that to transmit > information, as there is absolutely no way to influence the > final state of the photon after the measurement. > > But what happens if you have a stream of many such entangled photons ? > Quantum mechanics says that if you perform a measurement on a > system you modify its properties. So if i put a first > experiment on the path of the stream, and a second one a > little further down the stream, the second yields results > that are not identical to what you would get if the first > experiment was not there. In other worlds , the second > experiment knows that the first one is there. I can get two > different results by putting or removing the first > experiment, and that can be taken as 0 or 1. The problem being, since you don't know what the state was before the first experiment, how can you know whether a first experiment was added or not, by looking after the second experiment? All you know is that a first experiment *does* change the state. Nothing more. Bert ---------------------------------------------------------------------- You can UNSUBSCRIBE from the OpenDTV list in two ways: - Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word unsubscribe in the subject line.