Adaptive Finite Impulse Response Filter

This study develops a modified Sage-Husa adaptive filtering (SHAF) algorithm-based adaptive finite impulse response (FIR) filter for linear time-invariant systems. First, we modify the SHAF algorithm by applying the finite horizon sequential noise statistics estimation method and the FIR estimation scheme. Subsequently, we construct a novel adaptive FIR filter by combining the modified SHAF algorithm and the optimal FIR filter. Owing to the FIR structure, the developed method exhibits good properties, such as bounded-input-bounded-output stability, robustness, and adaptability to temporary model uncertainties. Based on computer simulations, the performance and effectiveness of the developed algorithm are demonstrated by comparing the estimated results with those obtained using Kalman, Sage-Husa adaptive Kalman, and optimal FIR filters.