Research On Array Source Enumeration Algorithm For Low Signal-to-noise Ratio

Most of the existing high-resolution array signal processing algorithms have the premise that the source number has been known or pre-estimated, which, however, is unknown in reality. And the deviation between actual source number and estimated source number may make the performance of many direction-of-arrival estimation algorithms deteriorate rapidly. Most existing source enumeration techniques in sensor-based studies have a satisfactory performance in high or middle signal-to-noise ratio(SNR), but almost lose effectiveness at low SNR. As a consequence, this paper focuses on the study of source enumeration algorithms regarding low SNR.Firstly, the basic knowledge of array signal processing is introduced and the classical narrowband source enumeration algorithms are studied including the eigendecomposition method, the information theoretic criteria, the spatial smoothing method and the Gerschgorin disk method. The algorithms are analyzed in theory, compared and simulated on algorithms’ performance.Secondly, considering the delay differential characteristic of the sensor elements in the uniform linear array, the construction and characteristic of the Hankel matrices of the spatial cross-correlation function of received data are introduced. And a source enumeration is proposed based on the singular vector of Hankel matrices for low SNR, which employs the Euclidean norm of the product vector of the steering matrix and left singular vectors of Hankel matrix. The proposed method utilizes the blind beamforming technique to estimate the steering matrix and the spatial smoothing scheme to dispose the dimension mismatched problem. Simulations are performed to display the better enumeration performance at low SNR situation and the validation for the non-coherent and coherent source signals.Finally, classic wide-band source enumeration algorithms are intruduced including the eigenvalue grads method of the incoherent signal subspace method(ISM) and the two-side correlation transformation method of the coherent signal method(CSM). And a wideband source enumeration method is proposed with the combination of Hankel matrix of the cross-correlation function based on the ISM idea. Simulations indicate the proposed method realizes the wide-band source enumeration, has the robustness and a favorable decorrelation capacity in low SNR situation.