MICROWAVE REPEATER RESEARCH 193 



Other observations on the New York-Neshanic microwave path may be 

 summarized as follows: While all the wavelengths were affected at tunes 

 of anomalous propagation, the shorter wavelengths faded more severely 

 and the character of the fading was different from that observed at the 42 

 centimeters wavelength; apparently the 3-10 centimeter range was more 

 sensitive to the fine structure of the atmosphere, as pointed out previously. 

 During non-fading periods, signal levels were very close to the free-space 

 values with the exception of the 1.25 centimeter signal which was usually 

 15 db or more below the free space value because of atmospheric absorption 

 effects. Some special tests showed that fading was considerably more severe 

 when one of the terminals was lowered so that the transmission path was 

 grazing slightly below line-of-sight. It was also found that fading was about 

 twice as great, in decibels, on the whole path as on either half-section. A 

 statistical analysis, on an hourly basis, of all the data on 6.5 centimeters 

 showed that only one-half of one percent of the total hours had signal 

 minima deeper than 20 db below the free space field. Also during August 

 1, the day of the most severe fading, the signal was more than 20 db below 

 free space for about one-half of one percent of the time. It was also found 

 that signals of the order of 10 db above free space were equally probable. 

 From a consideration of these statistics, it was decided to engineer the New 

 York-Boston repeater circuit with — 20 to + 10 db allowance for fading 

 on each link. 



Specialized Experiments 



Much of our more recent work on microwave propagation has been of a 

 specialized nature in which apparatus and experiments have been designed 

 more for the purpose of studying the mechanism of anomalous propagation 

 than for making a statistical analysis of the transmission. Perhaps the most 

 informative experiments have been those in which narrow beam scanning 

 antennas were used to explore the incident wave fronts. 



The first of these antennas had an aperture of 20 feet and a beam width 

 between half-power points of | degree at the design wavelength of 3.25 

 centimeters. It was built for the purpose of establishing a practical limit 

 to the size, and hence the directivity, of microwave repeater antennas from 

 the standpoint of variations in the angle of arrival of the received wave. 

 It had been realized, of course, that variations in the refractivity of the 

 atmosphere would cause some deviations in the path of the wave. While 

 these deviations should be negligible in comparison with the beam width 

 of antennas normally used in the ultra-short-wave region, they might be 

 comparable with the beam widths readily obtainable in the microwave 

 region. 



