White Noise Generation Using BOX_MULLER Algorithm and Simulation of Ship Radiated Noise

In signal processing and acoustic simulation domains, generating high-quality white noise serves as a fundamental step for modeling realistic environmental noise. The BOX_MULLER algorithm, a classical and efficient random number generation method, transforms uniformly distributed random numbers into Gaussian-distributed random variables. This transformation enables the generation of white noise signals with proper statistical characteristics through code implementation that typically involves trigonometric functions and logarithmic operations on uniform random inputs.

White noise exhibits a flat frequency spectrum where energy is uniformly distributed across all frequency components, making it a benchmark for noise modeling. The Gaussian white noise produced by the BOX_MULLER algorithm can be further enhanced to simulate ship radiated noise characteristics. Ship radiated noise typically comprises multiple components including mechanical vibrations and propeller cavitation. During simulation, filter functions can be applied to adjust the spectral properties of the white noise, making it statistically consistent with actual ship noise patterns. This involves implementing digital filters (such as IIR or FIR filters) to shape the noise spectrum according to empirical data.

This methodology not only applies to acoustic simulations but can also be extended to noise modeling in radar and communication systems. The generated noise provides a reliable background for subsequent research activities like signal detection and feature extraction algorithms, where proper noise modeling is crucial for testing algorithm robustness.