Multiple signal detection based on spatial-frequency adaptive processing using fast subspace decomposition method

As the positioning accuracy of GNSS systems will degrade in the presence of electromagnetic interference, jamming detection becomes pivotal part in high-precision wireless communication systems which is conventionally accomplished by energy statistics or time-frequency transformation. In this paper, we present the spatial-frequency adaptive processing (SFAP) method which will achieve highly detecting resolution through the distribution of eigenvalues in frequency domain, and compared to the conventional methods the SFAP can accurately detect the bandwidth of multiple signals. The application of fast subspace decomposition (FSD) using Lanczos algorithm iteratively implements the eigenvalue-decomposition with lower complexity while the source enumeration is estimated simultaneously. Moreover, the proposed method still works well even in the presence of weaker clutter or stochastic errors. Computer simulation results confirm the robustness of the detecting method.

I. Introduction

As satellite-based navigation systems, Global Navigation Satellite Systems (GNSS) have found applications in various sectors, but the problem that it is prone to be interfered by intentional jamming attracts much attention. Thus, there have been a multitude of researches on jamming detection, adaptive array processing comes into being accordingly, which can achieve high resolution in jamming detection. There are manifold techniques to detect interference in information theoretic statistics or time-frequency transformation [1]–[3], but among the conventional methods it is inevitable to increase computational complexity and disable to detect weaker clutters, and furthermore, it is impossible to distinguish bandwidth of multi-sources radiated in same center frequency. As the demand of high performance of jamming detection, the robust scheme based on spatial-frequency adaptive processing (SFAP) is to be analyzed and examined.