In this paper, the fuzzy mathematical programming problem is formalized based on the idea analogous with the chance constrained programming problem. The difference in meaning between the ambiguity of the coefficients and that of the decision maker's preference is emphasized. The fuzzy relations between possibility distributions proposed. by Inuiguchi et al. are introduced. The constraints with fuzzy coefficients are treated as the restrictions which should be satisfied properly rather than perfectly. Namely, the constraints are satisfied to at least a certain level given by the decision maker. The objective function with fuzzy coefficients is treated variously depending on the decision objectives, i. e. the optimization of the modalities, the optimization of the fractile and the minimization of the ambiguity. The determinstic equivalent constraints and the equivalent problems are shown when the constraints and the objective function are linear.
This paper is concerned with a digital learing control method in which the present input is modified by the error information at the previous trial to produce a given desired output. First, it is clarified that the input saturation and disturbances cause serious problems concerning to the error convergency and achievable accuracy in the digital learning control system, especially when the sampling period becomes small. Second, a digital learning control method is proposed which overcomes the above problems. Finally, the validity of the proposed method is illustrated by simulation and experiments.
In this paper, applications of the circular lattice (CL) method for multichannel signal processing to ARMA modeling are presented. After a review of the CL method, the covariance characterization of an ARMA process is discussed by embedding an ARMA process into a degenerate two-channel AR process. The connections with the method of Mullis-Roberts and the theory of Markov-cover by Anderson and Skelton are pointed out. Then, a new ARMA lattice filter is proposed and the effect of coefficient quantization on the frequency response of our lattice filter is compared with those of other ARMA lattice filters. Finally, an ARMA parameter estimation problem based on the impulse response data is treated by using the Burg type CL algorithm. Some numerical results are presented to show the effectiveness of the proposed method.
A practical method of power spectral estimation of stationary random signals is proposed, that can automatically provide an associated optimal estimate of the signals, only from an observed time series datum. First, the conventional method of power spectral estimation by smoothing periodgram is reviewed and the mean squared error of the normalized estimate is made clear, from the viewpoints of its bias and variance. Secondly, based on the fundamental nature of the statistics in a mutually reverse relation with respect to the smoothing bandwidth, a systematic method not only of adaptively determining the bandwidth but also of obtaining the optimal spectral estimate is presented, by newly introducing the evaluation indices of the statistics only from an observed datum. Finally, both the fundamental character and the effectiveness of the proposed method are illustrated by the fundamental results of computer simulations and power spectral estimation of measured heart sounds and noises.