THEORETICAL FUNDAMENTALS OF PULSE TRANSMISSION t t i 



increase in the pulse interval, as represented by the second case in Fig. 

 33, the maximum displacement of the zero line would be half the peak 

 amplitude of the pulses. There would then be a 50 per cent reduction in 

 the margin for distinction between the presence and absence of a pulse 

 in a random pulse train, rather than a complete elimination of the margin 

 for an infinite train of pulses of the same polarity transmitted at intervals 

 n = 1,-/1 • This impro\'ement would be achieved at the expense of a 

 two-fold increase in bandwidth for a given pulse transmission rate. A 

 further improvement, for the same tAvo-fold increase in bandwidth, can 

 be achieved by "dipulse" transmission, as discussed below. 



In dipulse transmission a positive pulse followed by a negative pulse 

 in the next pulse position would be transmitted to indicate "on," and 

 a negative pulse followed by a positive pulse to indicate '"off," as indi- 

 cated in Fig. 34. There will then be a substantial reduction in the pulse 



(a) PULSE TRAIN WITH POSITIVE AND NEGATIVE IMPULSES 



(b) PULSE TRAIN WITH POSITIVE AND NEGATIVE DIPULSES 



(Cj PULSE TRAIN (8] REVERSED AND DELAYED BY ONE PULSE INTERVAL 



2 3 4 5 6 7 



9 10 H 



1 — r 



J— I 



(d) DICODE TRANSMISSION (a) + (C) 



THE PULSES AND ZEROS IN THE RECEIVED PULSE TRAIN (cj ) HAVE THE 

 FOLLOWING RELATIONS TO THE ORIGINAL PULSES (3) 



1. POSITIVE AND NEGATIVE PULSES IN (dj REPRESENT CORRESPONDING 



PULSES IN (a) 



2. POINTS ON PULSE TRAIN IN Cd) REPRESENT A REPETITION OF PREVIOUS 

 PULSE, AS INDICATED BY DASHED LINES 



Fig. .34 — Dipulse and dicode pulse transmission methods. 



