PROPAGATION STUDIES AT MICROWAVE FREQUENCIES 103 



beyoiul this point results only in moving tiic pulses fartiier apart and has 

 no effect upon pulse shape or amplitude. 



The experimental set up just described proved to l)e somewhat un- 

 satisfactory since it was not possible with it to produce phase opposition 

 between the two carriers without having zero delay difference between the 

 two paths or a difference of at least 0.(i() millimicroseconds. For the length 

 of pulse used this latter amount of delay difference is sufficient to pre\ent 

 anything like complete cancellation of pulses. In fact the amplitudes of 

 the two i-esultant peaks are only about 12 db below the peak amplitude 

 of the original pulse. From this we know that for the natural path any 

 fade which appeared to be complete must have resulted from path dif- 

 ferences of less than 0.60 feet, in fact from differences of less than about 

 one-half foot. 



SUMMARY 



The pulse experiment results indicate that over one particular path 

 at least there is, at times, transmission of microwaves by at least two, and 

 probably more than two, paths. Path differences involved are from a 

 fraction of a foot up to about seven feet, differences of less than about 

 three feet being the most common. These results agree with those ob- 

 tained by other methods. These multi-path effects result in Imd distortion 

 of very short pulses and even in the presence of entirely separate spurious 

 pulses. These effects put a definite lower limit on pulselength and spacing 

 between pulses in a pulse transmission system. The limit depends upon 

 the amount of distortion which can be tolerated and also upon the per- 

 centage of time such distortion can be accepted. No statistical data were 

 recorded. 



With the lalioratory eciuipment for simulating transmission over two 

 paths, many of the waveforms obtained over the natural path could be 

 duplicated. There were times, however, when the waveforms received by 

 way oi the natural path were too complicated to be explained by trans- 

 mission by as few at two paths. 



ACKNOWLEDGEMENTS 



Space does not allow giving credit to all of the many people who con- 

 tri})Uted to the success of this project. A. F. Dietrich made the mechanical 

 layouts for most of the ecjuipment and supervised its construction as well 

 as assisting in taking data and in many other ways. I also wish to thank 

 J. C. Schelleng and W. 'SI. Goodall for guidance and suggestions. CI. M. 

 Eberhardt furnished the traveling- wave-amplifier circuits. 



