ABSTRACT
Gabor[J. Inst. Elec. Eng.PartIII,93(1946) argued that the usual statistical measures for standard deviation and mean in the frequency domain do not yield formulas in agreement with intuition when applied to the calculation of time and frequency dispersions for real signals. Gabor offered a prescription for correcting this problem which was to limit the summation in the calculation of frequency statistics to positive values and for consistency, to use the analytic signal for the calculation of time dispersions. A more recent statement of Gabor's prescription with applications can be found in L.Cohen, "Time-Freqency Analysis", Prentice Hall, 1995. The purpopse of this talk is to show that Gabor's procedures can generate divergent terms in the time dispersion formulas and values time-frequency dispersion products which are less then 1/4.