The BCGM-OR algorithm was derived on the basis of the results of error anal
ysis of the block orthogonal projection algorithm using the conjugate gradi
ent method. The authors have improved the convergence characteristics of th
is algorithm significantly under noisy conditions while at the same time re
ducing the computational load by optimizing the number of iterations for th
e conjugate gradient method. However, countermeasures are required for unst
able operation resulting from variations in the size of the input signal, w
hich is a fundamental limit of the orthogonal projection-type adaptive algo
rithm in the authors' method. With respect to the learning identification m
ethod when the block length of the block orthogonal projection algorithm is
1, the authors have already shown that: a guarantee value (the upper limit
of the estimated error after convergence) can be obtained by using a metho
d in which the coefficient updating is stopped when the state vector norm b
ecomes smaller than the set threshold value. Based on this method, the auth
ors first stabilize the guarantee value in the BCGM-OR algorithm when perfo
rming coefficient update cutoff. They then identify the threshold value use
d to satisfy the guarantee value and next describe the BCCM-OR algorithm wi
th a guarantee value. Finally, using computer simulations, the authors show
good convergence characteristics under noisy conditions for the proposed m
ethod, and then confirm that the operational stability is superior particul
arly with respect to nonstable input signals. (C) 2000 Scripts Technica.