300 



hELL SYSTEM TECHNICAL JOURNAL 



due to the dissimilar bandpass characteristics of the two channels of the 

 experimental unit used for the tests and part due to the lack of an extremely 

 sharp diversity selection. It should be pointed out that the conditions under 

 which the theoretical maximum diversity signal-to-noise condition may be 

 reached are very hard to obtain in practice. If the noise levels in the two 



15 5 10 15 5 



SIGNAL-TO- NOISE RATIO IN DECIBELS 



Fig. 38.-^Peak distortion versus signal-to-noise ratio characteristics for dual -diversity 

 operation with a common limiter. Signal-to-noise measured at output of 400 kc I.F. 

 amplifier. 



channels are not equal, and if the diversity selecting method does not at all 

 times exclusively select the channel with the greater signal, the distortion 

 characteristic will deteriorate accordingly. Because the AVC sensitivities 

 of two radio receivers may differ considerably the noise levels cannot be 

 maintained closely the same and usually no provision is made for determin- 

 ing the noise level except by ear. The slightly better diversity action which 

 can theoretically be obtained is therefore felt to be of doubtful usefulness 

 under actual op>erating conditions. The common limiter method has the 



I 



