Answers are grounded in this paper and cite their sources.
The author's previously published results (ibid., vol.8, p.1816-23, 1990) which dealt with the bit-error probability of a lightwave communication system whose noise contribution was exclusively attributed to spontaneous emission noise of optical amplifiers are extended. The earlier theory was based on an NRZ (nonreturn-to-zero) modulation format with zero signal power assigned to logical zeros. The extension is based on the same mathematical technique but includes Gaussian noise originating after the signal has been converted to an electrical current. This noise could be thermal noise, shot noise, or a combination of both. It is no longer assumed that the pulse representing a logical one fills its time slot completely. Now, the optical pulses can be much shorter than the bit period (solitons), the power level of logical zeros need not vanish, and the thermal noise of the postdetection electronic circuit is taken into account.>
No takes yet. Share an insight, caveat, or question.
D. Marcuse (1991) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: