Advanced Metric for Ultrasonic Signal Analysis Using Adaptive Dynamic Time Warping

Ultrasonic sensing plays a crucial role in various applications, including automotive safety, medical imaging, and industrial quality control.Simulators are increasingly used to replicate the behavior of ultrasonic sensors, reducing the need for physical testing.An essential challenge is to ensure the accuracy of the simulated signals compared to real-world measurements.In this work, we introduce a novel metric based on the corresponding signal processing pipeline and the Dynamic Time Warping algorithm to compare real and simulated ultrasonic signals.Unlike traditional distance metrics, it provides a flexible approach to align signals dynamically, compensating for variations in time shifts and distortions.Our method enhances this with an inner metric extended with gradient-based adjustments focusing on the characteristics of the ultrasonic signal.We evaluated the proposed approach using a dataset of real and simulated signals, demonstrating its superior performance in measuring signal distance and its applicability to validate simulator accuracy.The findings contribute to improved ultrasonic sensor validation methodologies, leading to more reliable and cost-effective sensor development.