Citation

Abstract

Analysis is presented for evaluating PPM error probabilities in optical photodetector receivers governed by Statistics due to Gaussian mixture densities. A Gaussian mixture density arises when a discrete variable (e.g., a photodetector count variable) is added to a continuous Gaussian variable (e.g., thermal noise). Making use of some properties of photomultiplier Gaussian mixture distributions, some approximate error probability formulas can be derived. These appear as averages of M-ary orthogonal Gaussian error probabilities, of which the latter are well documented in the literature. The use of a pure Gaussian assumption is considered, arid when properly defined, appears as an accurate upper bound to performance.

Details

Volume
42-68
Published
April 15, 1982
Pages
72–77
File Size
362.8 KB