Adaptive Subspace Detection of Spread Spectrum Signals With General Antenna Array Configuration

Assume that a low probability of detection communication (LPD) system has a generally configured antenna array receiver, comprising a primary array with a high-gain beam and a reference array with a low-gain beam. This paper addresses the signal activity detection of spread spectrum signals for such LPD systems in Gaussian disturbance with unknown covariance matrix. We assume that partial a priori knowledge about the direction of arrival (DOA) (i.e., the steering vector) of signals is available, and hence a subspace model is adopted to capture this information. Consequently, an adaptive subspace detector is proposed for the detection problem based on the Wald test criterion. Theoretical expressions of the detection probability and false alarm probability are derived for the proposed subspace Wald test in the general case, where the actual steering vector may not fall within the pre-designed subspace. It is found that the proposed subspace Wald test bears the constant false alarm rate property. Finally, Monte Carlo experiments are carried out, which validate our theoretical analysis. In addition, experiment results demonstrate that our proposed subspace Wald test outperforms the conventional ones that do not exploit the side knowledge of the DOA. It also reveals that the subspace Wald test demonstrates strong robustness against the DOA information inaccuracy.