Study On Interference Suppression For GPS Receivers

Global Navigation Satellite System (GNSS) is commonly deployed to measure the speed of a moving object and estimate its three-dimension position and time. Through increasing development and perfection, GNSS will be combined with internet and mobile network in the future and therefore applied to military and civilian areas widely. However, the GNSS signals are easily subject to intentional and malicious interference because they are inherently weak. How to guarantee the reliable and safe operation of GNSS in the jamming becomes more and more important. The interference for GNSS is mainly targeted at receivers on the earth and can be classified into overwhelming interference and spurious interference. In this thesis, we consider the issue of suppressing two kinds of overwhelming interference, i.e., FM interferer and wideband noise interferer on navigation receivers in the following aspects.First, we analyze the relationships of the Wigner-Ville distribution, Radon-Wigner transform and the fractional Fourier transform, and identify the feature of Linear Frequency Modulation (LFM) interference which is different from the navigation satellite signals'in the transform domain for a receiver with a sensor. Based on these observations, we propose a new approach to cancel LFM interference in the transform domain and, meanwhile, avoid any cross-term interference in the form of Wigner-Ville distribution.Second, in array signal processing we consider how to estimate the optimal weight vector in the strong jamming environment when any prior information of navigation signals is unknown. To solve this issue, we adopt the orthogonal projection and maximum signal-noise ratio technique to achieve the optimum weight vector, implement beamforming, and suppress interference and enhance desired signal.Third, the number of sources and the direction of arrival (DOA) of each source are two important parameters in array signal processing. The estimation of source parameters is usually executed under noise environment. Hence, the conventional approaches fail in the strong jamming case. In this thesis, we apply a combining technique to estimate the source parameters. More specifically, we use subspace technique to suppress interference, adopt transform Gerschgorin approach to exploit the received signal after interference cancellation for estimating the number of desired signals, and utilize the high resolution MUSIC method to identify the DOA of each desired signal.Finally, in the array signal processing, we investigate how to adapt a weight vector quickly for a changing jamming environment. In view of the excellent capability of the RLS algorithm in tracking an unknown parameter vector, the present thesis proposes an improving RLS-version algorithm based on the minimum output energy, derives the steps, and demonstrates its steady-state performance and tracking capability.