Robust DOA Estimation Using Modified VSSLMS for UAV-Assisted Disaster Management Applications
Аннотация
Accurate and timely Direction of Arrival (DOA) estimation is essential in Unmanned Aerial Vehicle (UAV)-assisted disaster management scenarios, where rapid localization of signal sources supports critical response operations. This paper proposes a robust and computationally efficient DOA estimation method based on a modified Variable Step-Size Least Mean Squares (VSSLMS) algorithm enhanced with a normalized sigmoid function for adaptive step-size control. The proposed algorithm dynamically adjusts the learning rate in response to the signal environment, improving convergence speed, tracking performance, and noise resilience in non-stationary conditions. Unlike traditional subspace-based methods, this approach eliminates the need for covariance matrix estimation and eigen-decomposition, significantly reducing computational complexity. Simulation results demonstrate the algorithm's superior performance in low Signal-to-Noise Ratio (SNR) environments and with limited snapshots, making it well-suited for real-time implementation on resource-constrained UAV platforms in emergency response missions.