78 THE BELL SYSTEM TECHNICAL JOURNAL, JANUARY 1952 



second. Since the intermediate fretiuency amplifier of the receiver is only 

 350 kc wide, (centered at 600 kc) narrow pulses are generated each time 

 the frequency of the transmitter crosses the freciuency to which the 

 receiver is tuned. These intermediate frequency pulses are displayed 

 vertically on a cathode ray tube. The horizontal trace is synchronized 

 with the one-second sweep rate of the beating oscillator. 



The normal daytime frequency-sweep pattern is shown in Fig. 7(a). 

 The vertical scale is linear in amplitude and the horizontal scale is almost 

 linear in freciuency, with frecjuency decreasing from left to right. Visible 

 at the extreme left is the signal used for continuous recording. Since 

 there is only one transmission path involved, the amplitude of the re- 

 ceived signal is nearly constant over the 450-mc band. If another signal 

 were present which had travelled over a path of different length, the two 

 signals would add when the frequency is such that the path length 

 difference is an even multiple of half-wavelengths and subtract when 

 the path length difference is an odd multiple of half-wavelengths. Simple 

 calculation shows that if the path length difference is one foot, the 

 frequencies at which the signals add and subtract are separated about 

 500 mc. Thus the limit of resolution for the frequency-sweep experiment 

 is a little more than one foot. 



Photographs taken on a night when the angle-of -arrival radar indicated 

 two almost equal components separated about 0.4 degrees in angle are 

 shown in Fig. 7(b). The time interval between the two pictures is 30 

 seconds, during which the minimum had shifted about 150 mc. The 

 pictures can be interpreted as simple two-path transmission with an 

 indicated path difference of about two feet and an amplitude ratio of 

 0.7 to 1. On this night the minimum shifted back and forth across the 

 frequency band — sometimes slowly and sometimes rapidly. At times the 

 position of the minimum might remain fixed but its depth would change. 



Photographs taken on a night when there were abnormal reflections 

 from the water of Raritan Bay are shown in Fig. 7(c). There are evidently 

 two main components with path difference of about six feet, with a small 

 third component causing the slight decrease in amplitude of the peaks 

 from left to right. These pictures were taken 9 minutes apart, but this 

 type of pattern was observed over a period of about three hours on this 

 night. 



Usually the frequency sweep patterns are considerably more compli- 

 cated than the ones shown so far. Fig. 7(d) shows two photographs which 

 indicate that at least three signal components and perhaps more were 

 present. The time interval between the two pictures was about 30 

 seconds. 



