Rough neural computing in signal analysis

This paper introduces an application of a particular form of rough neural c omputing in signal analysis. The form of rough neural network used in this study is based on rough sets, rough membership functions, and decision rule s. Two forms of neurons are found in such a network: rough membership funct ion neurons and decider neurons. Each rough membership function neuron cons tructs upper and lower approximation equivalence classes in response to inp ut signals as an aid to classifying inputs. In this paper, the output of a rough membership function neuron results from the computation performed by a rough membership function in determining degree of overlap between an upp er approximation set representing approximate knowledge about inputs and a set of measurements representing certain knowledge about a particular class of objects. Decider neurons implement granules derived from decision rules extracted from data sets using rough set theory. A decider neuron instanti ates approximate reasoning in assessing rough membership function values gl eaned from input data. An introduction to the basic concepts underlying rou gh membership neural networks is briefly given. An application of rough neu ral computing in classifying the power system faults is considered.