Detection and distance estimation of moving rotorcraft based on acoustic attenuation

Accurate acoustic detection and ranging of rotorcraft in low-altitude airspace is crucial for enhancing situational awareness and supporting collision avoidance in environments with drones. This study proposes a robust three-step approach for practical rotorcraft detection and distance estimation using real-world acoustic data. The first step applies the fast spectral coherence (Fast-SC) method to measured signals, generating a 2-D frequency–cyclic frequency matrix to emphasize periodic features from rotor blades. The second step employs a periodicity detection algorithm to extract blade-pass frequency (BPF) components, enabling reliable rotor signature identification even under drone and environmental noise. The third step estimates distance by matching the measured low-frequency power spectral density (PSD) to reference data, minimizing mean squared error (MSE) using the ISO 9613-2 model. For validation, on-board GPS ground truth data provide accurate position references. The proposed method is lightweight in terms of both computational resources and system mass, supporting real-time implementation. Experimental results with field recordings demonstrate that this Fast-SC-based framework enables robust, practical detection and ranging of rotorcraft, improving acoustic situational awareness for safe drone–rotorcraft coexistence.