CRITERIA FOR SYSTEM IDENTIFICATION WITH QUANTIZED DATA AND THE OPTIMAL QUANTIZATION SCHEMES

In this paper, we first examine several criteria for system identification with quantized output data and show that the ordinary parameter estimator for quantization-free case is still reasonable according to those criteria. Then, we give the optimal quantization schemes for minimizing the estimation errors under a constraint on the number of the quantized subsections of the output signals or the expectation of the optimal code length when the quantized data is encoded.