Real-Time Neuro-Fuzzy Digital Filtering: a Technical Scheme

In this paper, we describe the neural fuzzy filtering properties in real-time sense; giving an approach about the real-time neuro-fuzzy digital filters, defined in acronym form as RTNFDF. This kind of filters require an adaptive inference mechanism into the fuzzy logic structure to deduce the filter answers in order to select the best parameter values into the knowledge base (KB), actualizing the filter weights to give good enough answers in natural linguistic sense. The process requires that all of the states bound into RTNFDF time limit as a real-time system, considering the Nyquist criteria. The paper shows how to characterize the membership functions into the knowledge base in a probabilistic way with respect to the rules set decisions without loss of its real-time description, performing the RTFNDF. Moreover, the paper describes in schematic sense the fuzzy neural net architecture into the filter description. This kind of filters infer from different variables related of a reference system operation at its respective operation levels, classifying it responses in order to select dynamically the best answer for the infimum error limited by the error functional, respect to each variable considered. The results expressed in formal sense use the concepts exposed in the papers included into the references. Finally, we present in illustrative manner the RTNFDF operations using as a tool the Matlab software.