Stochastic fuzzy controller

Introduction: As we know, an implementation of fuzzy inference can be done on general purpose computer architectures (including digital signal processors and microcontrollers) or on a special fuzzy hardware [1]. Analogous and digital techniques are used. Fuzzy controllers working on stochastic logic represent a special kind of fuzzy hardware [2], [3]. The stochastic logic uses binary random signals with an average (mean) value in the interval between 0 and 1. It saves procedures known in the analogous technique, but it need digital compounds for its realization. It is possible to obtain a very simple realization of multiplication. The ways of signal representation are different. Synchronous or asynchronous working and various modulations are possible [4]. Computation with random signals has some advantages in comparison with classical techniques. Stochastic signals are much less sensitive to noise than analogous signals. Digital logic can be very space-efficient.