92 



THE BELL SYSTEM TECHNICAL JOURNAL, JANUARY 1952 



i.e., the path from Murray Hill, N. J., to Crawford's Hill (near Holmdel), 

 N. J.^ The path length is approximately 22 miles, and is partly over 

 water and partly over rough land terrain. The frequency sweep studies 

 had indicated that the path differences involved in multi-path fading 

 were of the order of one or two to about seven feet. In terms of delay 

 times this means differences of about 1 to 7 millimicroseconds. In order 

 to resolve the paths when the path differences were only one or two 

 feet, we should have liked to have pulses of about 1 millimicrosecond 

 duration. Because of the difficulties involved in generating, amphfying 



Fig. 1 — -Transmitting equipment. 



and detecting such short pulses, we accepted pulses which, when dis- 

 played on our final indicating equipment, had a length of 3 millimicro- 

 seconds at half ampUtude. (About 6 millimicroseconds at the base.) In 

 free space this pulse would be just about 6 feet long at the base. 



TRANSMITTING EQUIPMENT 



The transmitter was mounted on top of a 100-foot tower at Murray 

 Hill. As can be seen from Fig. 1, it consisted of a c-w reflex oscillator 

 operating at 4000 megacycles, a baseband pulse generator, a modulator, 

 or gate, for modulating these pulses on the microwave carrier, a single 

 stage traveling-wave amplifier and finally a horn antenna. Approximately 

 one watt of power was obtained from the transmitter at the peaks of 

 the pulses. The antenna area was 25 square feet and its gain 32 db above 

 that of a dipole. A pulse repetition frequency of 10 mc was employed. 



' A. B. Crawford and W. C. Jakes, Jr., "Selective Fading of Microwaves," 

 Bell System Tech. J., 31, Jan. 1952, pp. 68-90. 



