The Separation System of the Speech Signals Using Kalman Filter with Fuzzy Algorithm

The frequency reuse technology is used to supply a high system capacity in mobile communication system. However, the situation of the frequency reuse leads to co-channel interference. In this paper, we present an algorithm based on Kalman filter with fuzzy theory which separated the speech signals under additive white Gaussian noise channel. It is shown how to search the optimal weight coefficients value for minimizing mean square error (MSE) and high convergence time in the discrete state-space procedure. Then, comparing the traditional Kalman filter algorithm, we observe some disadvantage such as low convergence time and large mean square error. We adopt Kalman algorithm combined with fuzzy theory for improving the convergence speed and MSE of conventional Kalman filter. In numerical analysis, we discuss the convergence rate for carrier signal-to-noise ratio (SNR)_C and compare the MSE. Furthermore, in this paper the Kalman algorithm with fuzzy logic is better than Kalman filter performance for separating speech signal under additive white Gaussian noise (AWGN) channel