260 BELL SYSTEM TECHNICAL JOURNAL 



terminal above a flat roof that was approximately the same height as 

 other nearby flat roofed buildings, the field strength is more nearly 

 proportional to the height above the roof than to the height above the 

 ground. When the terminals are lowered below the average building 

 height additional complications are introduced. While this is some- 

 what difficult to picture because of the irregularity of the surface 

 bounding the transmitting medium ^^ the main outline seems simple 

 enough. Above the building level there is a tendency for the field to 

 follow the simple rules that hold for transmission over level country. 

 In general the field strength actually received in the street would be 

 proportional to the field strength overhead but of smaller amplitude 

 since it is a product of scattering. This does not imply that the street 

 signal comes down vertically; it probably is the result of scattering 

 from points lying in a fairly large zone about the receiver and consists 

 of a multiplicity of signals traveling in inclined directions. 



Returning now to the present data on the propagation of ultra- 

 short waves through urban areas, Fig. 2 shows ^^ that the field strength 

 is in general inversely proportional to the square of the distance from 

 the transmitter. The mean curve through the data is 12 db below the 

 curve for level terrain free from obstacles, plotted from equation (5) 

 above, indicating the additional attenuation due to man-made 

 structures. An analysis of the individual points shows that the 

 reduction in field due to the obstacles (i.e., in addition to the level 

 terrain attenuation) is independent of the distance so that there is no 

 absorption due to the buildings in the usual meaning of the word ; 

 otherwise the additional attenuation would increase with the distance. 



This method of interpretation is radically different from that of 

 investigators of the propagation of ultra-short waves through urban 

 areas whose papers have come to the attention of the authors."- ^^' '* 

 They have assumed that the transmission occurs as in free space 

 except for an additional attenuation through the absorbing layer of 

 buildings. Such an assumption that the propagation of ultra-short 

 waves is unaffected by the presence of the ground except in so far as 

 the waves penetrate the absorbing layer of buildings, appears to be 



" This surface is, of course, that formed by the ground and the walls and tops of 

 buildings. 



'- It will be shown later that the empirical formula assumed by Schroter, Sohne- 

 mann, Jones, and Muyskens aiid Kraus cannot be made to fit these data. 



" F. Schroter, "Zur Frage des Ultrakurzwellen-Runkfunks," E. N. T., \'ol. 8, 

 pp. 431-436, October, 1931. 



" K. Sohnemann, " Feldstarkemessungen im Ultrakurzwellengebiet," E. N. T., 

 Vol. 8, pp. 462 467, October, 1931. 



1"^ Henry Muyskens and John D. Kraus, "Some Characteristics of lUtra-High- 

 Frequency Transmission," Proc. L R. E., Vol. 21, pp. 1302-1316, September, 1933. 



