Active Noise Control Using Filtered X Lms — Simulation & Power Modeling

Abstract — Active Noise Control Using Filtered X Lms

Creative Commons Attribution (CC BY) license ( https://creativecommons.org/licenses/by/4.0/ ). This paper introduces a reduced Gaussian kernel filtered-x least mean square (RGKxLMS) algorithm for a nonlinear active noise control (NANC) system. This algorithm addresses the computational and storage challenges posed by the traditional kernel (i.e., KFxLMS) algorithm. Then, we analyze the mean weight behavior and computational complexity of the RGKxLMS, demonstrating its reduced complexity compared to existing kernel filtering methods and its mean stable performance. To further enhance noise reduction, we also develop the historical error correction RGKxLMS (HECRGKxLMS) algorithm, incorporating historical error information. Finally, the effectiveness of the proposed algorithms is validated, using Lorenz chaotic noise, non-stationary noise environments, and factory noise.

Analytical & State-Space Control Methodology

Keywords: nonlinear active noise control, kernel filtered-x least mean square algorithm, error-correction learning, nonlinearity issues Figure 2: Power Converter Circuit Schematic & Electrical Interfacing for Active Noise Control Using Filtered X Lms Based on the superposition principle, active noise control (ANC) technology utilizes loudspeakers to produce anti-phase sound waves with equal amplitude to cancel out unwanted noise [ 1 , 2 , 3 , 4 ]. However, there are some nonlinearities in ANC systems, such as distortion of the output signal and saturation nonlinearity if the loudspeakers and microphones are exposed to sound pressure levels above their tolerance limits. Furthermore, the reference signal may include nonlinear components, such as chaotic systems like Logistic chaotic noise , Henon chaotic noise, and Lorenz chaotic noise. As a result, linear ANC systems that utilize the filtered-x least mean square (FxLMS) algorithm might experience reduced effectiveness in noise reduction.

Simulink Waveforms & Power Quality Findings

In recent decades, a number of nonlinear filters have been researched [ 7 , 8 , 9 , 10 ] and developed to address various algorithm increases exponentially with the rise in filter order. Like the VF algorithm, the functional link artificial neural network (FLANN) is likewise a linear-in-the-parameters (LIP) network that uses various orthogonal basis functions, including Trigonometric , Chebyshev , Legendre , and Fourier . However, FLANN filters exhibit poor noise reduction kernel FxLMS (KFxLMS) algorithm, as proposed in demonstrates excellent nonlinear processing capabilities. However, its memory and computational complexity increase significantly with more adaptive iterations. To curb the growth of the kernel method network, various methods have been proposed, including the random Fourier filter (RFF) method, which restricts the input to a fixed-dimensional RFF space, and the Nyström (NM) method, which approximates the Gram matrix of the original kernel by a low-rank matrix. For the RFF, the accuracy of the approximation kernel is influenced by both the number of fixed dimensions and the frequency parameter of the random Fourier features. The NM method necessitates prior knowledge to generate a low-rank matrix that follows the same distribution as the reference signal. In addition to traditional nonlinear filters, deep learning-based ANC systems have emerged as a powerful alternative, leveraging neural network architectures and large-scale training datasets to capture complex nonlinear noise features [ 19 , 20 , 21 , 22 ]. For example, ref.

Scope & Practical Applications

• High-efficiency DC-DC converters and isolated resonant topologies (LLC / DAB)
• Sinusoidal and Space Vector PWM inverters for grid integration & industrial motor drives
• MOSFET / IGBT gate driver optimization with active Miller-clamp shoot-through protection
• Power quality conditioning with Total Harmonic Distortion (THD) strictly compliant with IEEE 519
• Complete Simscape .SLX model files, parameter calculation scripts, and IEEE dissertation reports