SIGNAL IDENTIFICATION DEVICE USING GENETIC ALGORITHM AND ON-LINE IDENTIFICATION SYSTEM

摘要

A feature parameter (primitive feature parameter) for recognizing the feature of a signal is defined, the primitive feature parameter is self-reorganized by a genetic algorithm (GA), and a new feature parameter (GA feature parameter) is automatically generated. If the result of evaluation of the GA feature parameter by using a distinction index or evaluation index is acceptable, a probability theory, the Dempster & Shafer probability theory, and a possibility function are applied to the GA feature parameter to generate a possibility distribution function and a judgment criterion for distinction.