On the applicability of the Fokker-Planck method in telecommunications

In the telecommunications area there are different opinions on the applicability of the Fokker-Planck (F-P) method. One group of researchers and engineers applies the F-P equation to processes which in terms of their probability density function (PDF) may be considered as a subset of the Markov processes, i.e. as (approximately) continuous Markov processes. Another group of researchers and engineers applies the F-P equation to physical processes irrespective of the amplitude distribution of the (noise) process which drives the system, as long as this driving process is of bandwidth considerably wider than the system bandwidth. The author focusses on the incompatibility of the two opinions which can be observed in the relevant synchronization literature. Examples, in particular synchronization systems, are considered and the potentially severe discrepancies are visualized as they can occur between actual system performances on one hand and analytical results from an unjustified application of the F-P method on the other hand.<>