As an adaptive equalization method for multipath fading in a high‐speed mobile communication at 100 kbit/s or more, a decision feedback equalization method with a Kalman filter is proposed based on the time‐variant model of the impulse response variation of the transmission path. In this paper, by means of this equalization method, the autoregressive (AR) model of the variation is derived from the measured values of the impulse response variations. The result of simulation of the equalization is described. First, a method is explained to derive an AR model to be set in the equalizer from a set of measured data of the impulse response. Next, the results of simulation are presented for the case in which the variation characteristics of the transmission path and the equalizer model correspond to each other. It is shown that the results have the error rate almost equal to the case of uniform fading without a delayed wave and are better than those from the LMS method. Further, since the variation of the transmission path is slower than the transmission speed, a model is introduced in which the pole of the AR model is approximated by 1. It is shown that the variation of the transmission path in a wide region can be equalized without significant degradation even if the equalizer model is fixed.
No takes yet. Share an insight, caveat, or question.
Ara et al. (1991) studied this question.