Read the paper by KA9Q and W3IWI that they gave at the 1996 Central States VHF Conference. As I read it, ham EME using advanced modulation techniques will likely be rather narrow-band. Most of the following applies to 100-200 W stations and "Oscar" beams. I'll quote a few things:
[regarding the EME propagation path] "... most signals have a coherence bandwidth ~500 Hz to 1 kHz ...." "... it also places an limit of ~300-500 baud on the signalling rates for EME digital communications." "At longer EME wavelengths (like 2m)...:" "10-20 cm/sec libration velocities mean that signals maintain coherence for several seconds." [Other refelction characteristics are relaticely unimportant.] "In contrast, at frequencies above 1 GHz, the 'rules' are different:" "10-20 cm/sec libration velocities reduce coherence times to a small fraction of a second."
"And the 70 cm band represents a transition region" "Since the typical Oscar satellite station is equiped for 2m and 70cm operation ... we envision it will be the 'home' [for] the signalling processing ideas we present here." "... we note that the bandwidth that can be used depends on three factors: > The added bandwidth due to spectral broadening described above. Physics makes this the ultimate limit!" > [frequency stability of the radios] > [ability to set the radios to compensate for Doppler effects]
[modulation] "All these factors become more serious [in] direct proportion to frequency. Schemes which may work at 2m or 70 cm will probably fail miserably at 10 GHz! The following section will develop a design which ends up with 64-ary FSK occupying a ~2 kHz bandwidth (i.e. channel spacing ~ 30 Hz), symbol rates ~2 baud, and some degree of coherent detection suitable for 2m and 70cm. At 13 cm or at 2.8 cm, we may have to use binary or 4-ary FSK, symbols several seconds long, ...." [but] "... the allowable ... symbol time is upper bounded by the channel coherence time (typically a second at 70 cm, several seconds at 2m), and lower bounded by the multipath delay spread (a few milliseconds at 2m and 70cm)" "The minimum symbol time is also bounded by the need to attain sufficient SNR to meet detector threshold. It is also bounded ... by how accurately we know the beginning and end of each symbol."
I thank Phil and Tom for letting me post the above without permission.
[With noncoherent detection reuired for the higher bands, longer symbols could be used.] [A short example of high power stations and a 300 bps data rate requiring 64 kHz RF bandwidth is also presented.]
So, if they are right, we won't see 500 kHz wide "SS" EME stations on the ham bands.
The W I D E SS users will be local terrestrial links, just down the street from you.
73 de Bob w3otc ------ Submissions: firstname.lastname@example.org Subscription/removal requests: email@example.com Human list administrator: firstname.lastname@example.org