716 



THE BELL SYSTEM TECHNICAL JOURNAL, JULY 1951 



voltage 7v, replace N by Nr, and (12) becomes 



— = ^' 



Nr 



I 

 %r 



(13) 



The physical significance of this ratio may be explained as follows. The 

 voltage which appears at the output of the system after the occurrence of a 

 PAM pulse may be represented as the sum of a noise voltage and a signal 

 voltage. Suppose that a very large number of measurements were made of 

 the noise voltage which occurs along with a preassigned value of signal 

 voltage. If the magnitude of the signal voltage is within the exponential 

 range of the expandor, then the ratio of signal voltage to the rms value 



15 



3 



UJ 



=)H -15 

 (0< 



«0CC-20 

 O 



z 



-30 



-25 



70 



65 60 55 50 45 40 35 30 25 20 15 10 5 



EXPANDOR OUTPUT VOLTAGE, E, IN DECIBELS BELOW 1 VOLT MAXIMUM 



Fig. 7 — Noise susceptibility. 



of observed noise voltages would equal SjlSIr in (13). The same value of 

 SjNr would be obtained if the measurements just described were repeated 

 for any other preassigned magnitude of signal voltage within the exponential 

 range of the expandor. 



When the voltages impressed upon the input of the expandor are within 



the linear range of the characteristic, the noise susceptibility, designated 53, 



17 

 is equal to -^ . From (7) we write 



^. = i 



(14) 



This shows that, within the linear range of the compandor, noise suscepti- 

 bility is inversely proportional to expansion ratio. 



