Correlation Analysis of Underwater Vehicle Self-Noise Collected by Microphone and Hydrophone

Underwater unmanned vehicles (UUVs) have gained significant attention in recent years due to their strong con-cealment, flexibility, high cost-effectiveness, and suitability for cluster operations. However, the self-noise of UUVs acts as a strong near-field interference source, severely affecting their detection capabilities and leading to a substantial decrease in sonar performance. To investigate the correlation between UUV self-noise collected by different methods under the same noise source, an experiment was conducted in an anechoic pool using a microphone and a hydrophone to simultaneously collect the self-noise of a UUV at different propeller speeds. The collected noise data were then compared and analyzed in terms of time domain, frequency domain, DEMON spectrum, and cross-spectrum. The results show that the noise collected by the microphone and hydrophone exhibits significant similarities in frequency distribution and a strong correlation in both time and frequency domains. The Pearson correlation coefficients at propeller speeds of 500r/min, 1200r/min, and 1800r/min were 0.6255, 0.8007, and 0.6105, respectively, indicating a strong correlation between the two types of noise. Furthermore, the cross-power spectral density analysis revealed obvious statistical correlation characteristics between the two noise signals in the frequency range of approximately 500Hz to 2000Hz. These findings provide a theoretical and practical basis for future research on noise mapping and spectrum subtraction cancellation techniques, which can enhance the passive detection capabilities of UUV sonar systems in complex underwater environments.