Detection of the number of coherent narrowband signals with L-shaped sensor array

This paper deals with the problem of estimating of the number of coherent narrowband signals impinging on the L-shaped sensor array, which is composed of two uniform linear arrays (ULAs) connected orthogonally at the one end of each ULA. An efficient detection method is proposed for the incident signals without the computationally burdensome eigendecomposition, where the number of incident signals is revealed as the rank of a combined matrix formed by the cross-correlations between the array data along two ULAs. Then a ratio criterion is presented in term of the row elements of QR upper-triangular factor of the autoproduct of the combined cross-correlation matrix, and the number of signals is determined as the value of the running index for which this criterion is maximized. The effectiveness of the proposed method is verified through numerical examples, and the simulation results show that the proposed method performs well in estimating the number of signals with small angular separation at relatively low signal-to-noise (SNR) and/or with a small number of snapshots.