Abstract :
In order to determine the relation between the output and input signal/noise ratios of a p.c.m. system, it is necessary to determine the probability of error in selecting a single pulse in a background of noise. This reduces to a problem in statistics which requires the use of a criterion for deciding as to which of two possible probability distributions a particular sample belongs. The ¿ideal observer¿ criterion is applied in this paper and the results are compared with those obtained by Flood1 Considering the difference in initial assumptions and method of solution, there is a surprising similarity between the results predicted by the two methods.
