Stochastic Model Predictive Control of Cheer Spikes and Low-Frequency Noise in Outdoor Festivals

This study addresses an open challenge in control systems—real-time suppression of random noise with jump and diffusion components—by proposing a method using Poisson- and Wiener-driven stochastic differential equations (SDEs) with model predictive control (MPC) for outdoor festivals. Achieving over 90% suppression of cheer spikes (1–5 kHz) and low-frequency noise (0.1–1 kHz), the approach introduces a novel extension of linear SDEs with Van der Pol and Duffing-type nonlinear SDEs, ensuring stability via a rigorously proven stochastic Lyapunov function. Detailed analyses of computational cost, robustness, and chaotic control, grounded in classical mathematical rigor, address real-time challenges. Visualization in time and frequency domains validates practical utility in acoustics engineering, offering a new paradigm for control systems. Recent studies have explored dynamical system responses under combined Poisson and Gaussian white noises, providing insights into their complex behaviors [14]. Furthermore, advancements in stochastic model predictive control have extended to handling sub-Gaussian noise, enhancing robustness in uncertain environments [7, 2, 5].