Curve-Fitting-Based FIR Filter Design for Stop-Band Attenuation Performance Improvement

Abstract

Finite impulse response (FIR) filters are widely used in signal processing due to their inherent stability, linear-phase property, and implementation robustness. However, achieving high stopband attenuation with a limited filter order and a steep attenuation slope remains challenging. This paper presents a curve-fitting-based FIR filter design method aimed at enhancing stopband attenuation performance. The proposed approach employs curve fitting to estimate the optimal parameters of a General Gauss1 model by fitting its functional form to a given impulse-response shape. The resulting mathematical model is then used to compute new FIR coefficients that enforce a smooth impulse-response profile, effectively suppressing oscillatory sidelobes. Consequently, the magnitude response exhibits a monotonic decay and an improved attenuation slope in the stopband. MATLAB simulation results confirm the effectiveness of the proposed method, demonstrating significant stopband attenuation enhancement without increasing the filter order.