DOA estimation of multiple sparse sources using three widely-spaced sensors

Traditionally, Direction Of Arrival (DOA) estimation methods require an inter-sensor spacing that does not exceed half the minimum wavelength of the impinging signal in order to avoid ambiguities. Placement of sensors at the minimum half wavelength is not physically realizable for some types of signals. Additionally, for some categories of signals, half wavelength spacing is not sufficient to achieve good DOA resolution. In this paper, our previously proposed delay disambiguation method is combined with a time-delay estimation method that exploits source sparsity in time-frequency domain to unambiguously estimate the DOAs of multiple sources impinging on three widely-spaced sensors. The approach is applied to DOA estimation of multiple ultrasonic Frequency Hopping Spread Spectrum (FHSS) sources in indoor environment. Experimental results emphasize the robustness of the approach even when the number of sources exceeds the number of sensors.