Study and Comparison of Interference Mitigation Techniques for GPS Receiver

V. Calmettes, F. Pradeilles and M. Bousquet

Abstract:	Over the last years, a large number of mitigation techniques have been investigated to improve the performance of the GPS receivers. The aim of this paper is to analyze and to compare different pre-correlation methods used to eliminate CW, FM and pulsed interference. The first technique consists of a pre-whitening linear filter in front of the conventional receiver optimized for a white gaussian noise. The filtering performed in frequency domain yields to large computing load. Moreover it can result in a significant correlation loss in the presence of wide band interference such as FM or pulsed jammers. Hence we suggested a structure made out of adaptive notch filters connected in series, and controlled by the LMS algorithm [1]. This structure is modified here to take into account the complex signal and to track multiple jammers. The second technique is derived from the detection theory. When the signal is buried in a non-gaussian noise a locally optimum test is defined. By expanding the probability density function (PDF) around the H0 hypothesis the likelihood ratio of the optimum receiver is obtained [2]. In that case the detector uses a non-linear function deduced from the P DF of the noise. This theory leads to a non-linear operator applied to the amplitude of the complex GPS signal defined by its polar components. This technique, which is known as Amplitude Domain processing (ADP)[3], can be implanted in the time or in the frequency domain. These filtering methods are presented and analyzed in this paper. The interference consists of sinusoidal signals such as CW, FM and pulsed waves. For each of them the processing gain, which gives the attenuation on the interfering power and the"cer~elation gain, which is a measure of the degradation on the GPS signal, are estimated. First a brief description of the conventional receiver is given to introduce the correlation and the processing gains that permit to assess these mitigation techniques. Then the linear filter is studied. Its performances are deduced from the analysis, at the convergence, of the statistics of the estimation error. They are determined, according to the type of the interfering signal. The next point concerns the analysis of the non linear filter. Here the performances depend on the operator applied to the amplitude of the signal. This operator is derived from the noise probability density function (PDF). Its expression is given in the presence of the different interfering signals. Then the processing and the correlation gains are deduced when the amplitude processing is carried out in the time domain. It is shown that the efficiency of this technique decreases with the number of jammers. For this reason a Discrete Fourier Transform is performed to separate the components of the interference and the amplitude processing is characterized in the frequency domain. Finally two solutions are proposed to eliminate multiple j ammers.
Published in:	Proceedings of the 14th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 2001) September 11 - 14, 2001 Salt Palace Convention Center Salt Lake City, UT
Pages:	957 - 968
Cite this article:	Calmettes, V., Pradeilles, F., Bousquet, M., "Study and Comparison of Interference Mitigation Techniques for GPS Receiver," Proceedings of the 14th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 2001), Salt Lake City, UT, September 2001, pp. 957-968.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In