Abstract: A brain-wave analysis method for analyzing a brain wave by extracting a power spectrum of the brain wave from detected brain-wave biological signals, detecting the presence or absence of peaks formed in the frequency band of each brain-wave rhythm in the power spectrum, and analyzing the brain wave based on the presence or absence of the peak spectrum; and brain-wave analysis apparatus for realizing the aforementioned method comprising a brain-wave extracting device for extracting a power spectrum with respect to a brain wave from detected brain-wave biological signals, and a peak detecting device for detecting the presence or absence of spectrum peaks formed in each brain-wave rhythm frequency band in the brain-wave power spectrum extracted by the brain-wave extracting device.

Abstract: A filter-coefficient conversion system relating to n pieces of digital filters for obtaining n pieces of frequency components in a desired frequency band using filter coefficients a(i, j) (where i=0, - - - m.sub.j, j=1, - - - n) described in a given filter-bank table, comprising a first transfer-function processing device for obtaining transfer functions H(a(i, j)) for the filter coefficients a(i, j) corresponding to each filter (j) in a desired frequency band, a variable-range generating device for generating gradually value-changing ranges RANGE (m), a second transfer-function processing device for obtaining new converted filter coefficients A (i, j) having rounded integer values based on the filter coefficients a (i, j) in the filter-bank table, preparing a filter-bank table thereof, and obtaining transfer functions H(A (i. j)) for the converted filter coefficients A (i, j), and an evaluating device for carrying out evaluation processing with respect to a given degree of tolerance.