A DIGITAL FILTER FOR ACTUAL STOCHASTIC SYSTEM OF MIXED TYPE BY USE OF THE HIGHER ORDER CORRELATION INFORMATION―ESTABLISHMENT OF THEORY AND ITS APPLICATION TO ACOUSTIC ENVIRONMENT

Abstract

In this study, a new type of successive state estimation method based on the Bayes' theorem is theoretically proposed for a mixed model type stochastic system with a nonlinear observation mechanism, especially in a quantized level form matched to the signal processing by use of a digital computer. Furthermore, the validity and the effectiveness of the proposed theory are experimentally confirmed by applying it to the actual environmental noise data in the room acoustics.