An underground twisted pair line is one mile long. The line is used to send a 1.50 Mbps signal as a polar binary NRZ voltage waveform. A logical 1 is transmitted as a +1.0 V pulse and a logical 0 is transmitted as a -1.0 V pulse. At the output of the receiver's sampling gate the pulses have peak amplitudes of +100 mV and -100 mV. Assume an ideal link and correct synchronization of the clock in the receiver.
a. What is the main cause of bit errors on this digital link?
b. At the receiver, what is the rule that is used to decide whether to output a 1 or a 0?
c. A logical 1 is transmitted over the link. A noise spike of +150 mV is added to the signal at the instant it is sampled by the receiver. Will this cause a bit error?
d. Observation of the voltage at the output of the sampling gate of the receiver shows that voltage spikes of 150 mV amplitude are being added to the signal pulses at an average rate of six spikes per second, with an equal number of positive and negative spikes. What is the bit error rate on the link?
e. What change at the transmitter would make this link error free?
(a) Dispersion over the link (i.e. the non uniform attenuation of frequency components that constitute the perfect (assumed) square pulse waveform) will both alter the waveform height (amplitude) and spread out the pulse energy (flatten the shape of the pulse) over the time interval. This dispersion effect will become more pronounced the further the pulse has to travel along the link. This results in pulse energy merging from one pulse into the period occupancy of an adjacent pulse. When the interfering pulse energy becomes great enough (i.e. equal or greater than the actual pulse energy to be measured in a time interval) then an ...
From a description of a digital link the bit errors on the link are deduced and analysed and questions on its QoS addressed