Fading Due To Multipath

I did some extensive testing about 30 years ago. The frequencies that your asking about is not the cause of multipath however. Absorption or reflection of these frequencies by physical objects is the dependant variable. The cause of multipath is per the definition of "multipath". The frequency (wavelength) is an independant variable. When RF is reflected, from the straight line path, the signals add (reflected + or - the direct signal). Which causes signal strength max and min of 1+1=2 or 1-1=0 db. And, any signal strength in between and is phase dependant. It's a simple trig problem which can be calculated by using the distance from the source to the reflector (trees, buildings, grain storage bins, etc) and the direct signal path to the receptor. If you have lots of reflectors and cannot place an antenna at a point that is without multipath then you will need 2 or more antennas and switch automatically between the antenna. Multipath is difficult when encountered in a moving environment such as a vehicle.

There are many aspects relative to fading due to multipath that include the bandwidth, type of receiver you are considering (rake taps), velocity of the mobile, etc. If you are looking for some place to start, I suggest getting ITU-R P.1546. It is also referenced in a recent article in the IEEE transactions on Vehicular Technology (Jan. 2008) titled, "Evaluation of the Propagation Model Recommendation ITU-R P.1546 for Mobile Services in Australia. If you are looking for more simple models consider using a Rayleigh (no line of site path) or Rician model (with a direct LOS path but you have to specify a K-factor). In both of those cases you will have to specify a set of path delays and a velocity/doppler shift as it affects the fading rate. P.1546 provides some of the parameters. You can also check the 3GPP standards (25.101, I think?) for channel models used to evaluate the performance of GSM. This is a complicated subject matter, and I hope this helps. Good luck.

FRESNEL MICROWAVE UHF PROPAGATION :

These words should yield volumes of empirical -- and often confused/confusing--data.