APPLICATION OF WIRE TRANSMISSION TO RADIO 129 



our old friend " static " or some interference is experienced 



from high frequency transients in power systems. Unless the 



Hnes are especially well transposed for these frequencies, the 



interference requires that the transmission level be kept above 



/i 

 a minimum value of the order of logio y = 1.2 (about — 25 



miles of standard cable below zero level). 



(b) While for radio telephone transmission the available data are 

 as yet very meagre, we have obtained a few order-of-magnitude 

 figures which should be of interest. For the Catalina Island 

 radiophone system, for example, the minimum field intensity 

 is estimated at roughly 1,000 microvolts per meter. The 

 circuit is sometimes quite noisy during the summer months 

 altho not prohibitively so. In our ship-to-shore radio tele- 

 phone experiments along the Atlantic coast, we have on occa- 

 sions worked with lower field intensities, as low as 100 micro- 

 volts per meter. The latter figure, however, gives a grade of 

 service far below wire standards. 



(c) The best data on the minimum permissible transmission level 

 for radio telegraphy are those obtained from the experience in 

 trans-Atlantic telegraph operation. The figures prevailing 

 for present trans-oceanic radio-telegraph operation are under- 

 stood to lie in the order of 10 to 100 microvolts per meter, 

 depending upon individual cases and the time of the year. 



The Net Transmission Equivalent 



The net over-all transmission equivalent of the system is measured 

 by the ratio of the transmitted to the received signaling power, and is 

 shown in Fig. 4 as the difference between the transmission levels at 

 the two ends. This relatively small loss represents the difference 

 between two large values, the transmission loss and the transmission 

 gain thruout the system. Relatively small changes in either the at- 

 tenuation or amplification may, therefore, cause large changes in the 

 net equivalent of the circuit, thus tending to give rise to instability 

 in the transmission performance of the circuit. 



This problem of fluctuation becomes very serious with the use of 

 very high frequencies, whether transmitted by wires or by radio. 

 Were we to attempt to employ, for example, a million cycles for wire 

 carrier transmission over considerable distances, as has been proposed, 

 not only would the losses be very large, but they would be unstable, 

 changing with weather conditions, so that the maintenance of a 



