BACKGROUND NOISE IN BROADCASTING 349 



as may occur in rural areas without causing the Hstener to abandon 

 attempts at reception, the flutter will be the more important. Conse- 

 quently, an improvement in the service in such regions would be ob- 

 tained by synchronizing the carriers of the two stations, even though 

 they continue to transmit different programs. 



Effect of Interfering Carrier on Desired Program 



Even if the interfering wave were unmodulated and there were a 

 neglibible noise background, there still remains the possibility of dis- 

 tortion of the desired program by the heterodyning action of the un- 

 desired carrier. In order to determine how important this effect is as 

 compared with those which have been discussed, a modulated carrier 

 (derived from WABC) and a w^eaker unmodulated wave were used. 

 A beat frequency of about 3 cycles was maintained during the course 

 of these observations. 



With the linear rectifier it was found that a perceptible distortion 

 of the desired program could not be detected on speech and jazz music 

 until the weak carrier was brought within 1 db of the strong one. 

 When the program consisted of music containing many sustained notes, 

 such as occur in a violin solo and even in vocal solos, the cyclic varia- 

 tions in output level were more noticeable. In such a case a ratio 

 of about 4 db was necessary to reduce the distortion to the detectable 

 limit. 



With the square law rectifier it was found that a carrier ratio of 

 10 db produced detectable distortion with any type of program. At a 

 ratio of 16 db distortion could be detected only when the program con- 

 tained sustained notes, and at 18 db could be noticed only when the 

 notes were sustained for a considerable time. 



The dependence of the permissible ratio upon the type of program 

 led us to make a similar observation when the strong carrier was modu- 

 lated 30 per cent with a pure tone of 400 cycles. Under such condi- 

 tions it was necessary to reduce the interfering carrier to about 34 db 

 below the strong one before the 3-cycle variation in the pure tone 

 definitely vanished. 



Conclusions 



The studies which have been reported furnish quantitative data on 

 the various types of interference which are encountered in shared 

 channel broadcasting and show what relative levels of interfering 

 carrier may be tolerated under various conditions. 



In high grade service areas the program from the undesired station 

 will be the most serious form of interference, provided the carrier beat 



