Detection of the number of sources at low signal-to-noise ratio

A new method to detect the number of sources in array signal processing is proposed. Most source detection techniques perform reasonably well at medium or high signal-to-noise ratio (SNR), but not at low SNR. This method exploits eigenvectors, instead of sample eigenvalues, for source enumeration. It employs the blind beamforming technique and the peak-to-average power ratio based frequency estimation algorithm to estimate the number of sources. Simulation results show that the proposed method is superior to the minimum description length and predictive description length algorithms at low SNR.