210 



ANGLE-OF-ARRIVAL EXPERIMENTS 



function ol tilt' angle cz at the receiver. In the case of 

 the angle a = 0.355° (deviation from true bearing 

 equal to -|-0.466°), the calculated index at the level 

 of total refraction, which computes as 505 ft, is very 

 closely that observed at the uppermost level of mete- 

 orological sounding (503 ft). Thus an angle of ar- 

 rival deviating by as much as 0.47° from true geo- 

 metric bearing is entirely possible for the meteoro- 

 logical situation of 0800, July 7, 19-±-i and for the 

 positions of the New York transmitter and Beer's 

 Hill receiver and could have been predicted from the 

 observed meteorological sounding. 



A second significant conclusion can be readily de- 

 duced by considering the modified refractive index 

 distributions ]-ec|uire(l in the layer immediately above 

 the transmitter for diiferent values of a. It is ap- 

 parent that the lapse of modified index required for 

 any of the angles considered in this example is not 

 substantially different among all four angles; the 

 primary requisite for the larger a's is that the lapse 

 continue t» greater heights. Thus relatively small 

 fluctuations in the meteorological elements can cause 

 a time change of 0.1° in the angle of arrival measured 

 at the Beer's Hill receiver. Furthermore, a particu- 

 larly unfavoraljle combination of small changes iu the 

 meteorological elements in this layer may cause the 

 signal at the receiver to fall to a very low level. A 

 similar conclusion is not valid for the case of propaga- 

 tion confined to the layer between transmitter and re- 

 ceiver and the case of path penetration below the re- 

 ceiver, since another slightly difl:'erent path can al- 

 ways be found along which energy can reach the re- 

 ceiver directly. 



A third significant conclusion can be deduced by 

 inspecting the comjjuted deviations from true bear- 

 ing of the angles at the receiver and the transmitter, 

 as given iu Table 3. It will be noted that the devia- 

 ti(jn from true bearing of the angle of arrival at the 

 receiver is not the same as the deviation from true 

 bearing of the angle of departure at the transmitter. 

 In fact the data of Table 3 indicate that, under the 



O 0.1 0.2 0.3 0.4 0.5 0.6 0.7 



DEVIATION OF yS AT XMTR FROM TRUE BEARING 



FiouRE 7. Cori'elation of deviations. 



meteorological situation of 0800 on July 7, 1944, the 

 angle of arrival at the receiver would have been 

 -(-0."i7° aljove true bearing in place of -|-0.46°, were 

 the receiver and transmitter interchanged at the end 

 jioints of the path. On the other hand, for both cases 

 1 and 2 treated above, the meteorological stratification 

 was such tliat the angular deviations from true bear- 

 ing at both receiver and transmitter were the same. 

 The relations of the angular deviations from true 

 bearing at the end points of the i^ath are summarized 

 in Figure 7. 



A fourth conclusion can be deduced l)y considering 

 the curve in Figure 7 based on the computations tab- 

 ulated in Talile 3. It will be recalled that a fluctuation 

 of 0.1° in angle of arrival at the receiver was con- 

 eluded as possible as a result of relatively small fluc- 

 tuations in the meteorological elements in the layer 

 immediately above the transmitter. But it should now 

 be noted that fluctuation of angle of departure at the 

 New York transmitter is approximately 0.35° when 

 the Beer's Hill angle of arrival varies approximately 

 0.1° for the particular meteorological situation of 

 0800 on July 7, 1944. 



It therefore follows, in summary, that the deviation 

 from true bearing measured at the position of the re- 

 cei\er depends not only on the range between trans- 

 mitter and receiver and on the meteorological condi- 

 tions but also and equally well on the relative differ- 

 ence iu heights of transmitter and receiver and the 

 position in height of the receiver with respect to the 

 transmitter. 



