Spread-spectrum Navigation Signal Detection In Non-gaussian Channel

As the electromagnetic environment becomes increasingly complex, spread spectrum navigation receivers are vulnerable to the growing problem of non-Gaussian interference, while the domestic research on the suppression of non-Gaussian interference has just started. This paper mainly investigates non-Gaussian interference suppression technique in spread-spectrum navigation system and proposes a signal detector based on frequency domain amplitude processing techniques. Noise observation in spread-spectrum navigation system is modeled as Alpha stable distribution. Based on the Generalized Central Limit Theorem, symmetric Alpha stable distribution is not only a very good model for atmospheric noise description, but also describes the non-Gaussian and heavy-tailed nature of signal statistical distribution.This paper applies locally optimal detection technique to the spread-spectrum navigation system to acquire a new signal detector, in which the noise model is modeled as symmetric Alpha-stable distribution. Then in different noise ratio and Alpha noise figure condition, conventional square detector and detector based on locally optimum detection are simulated and compared to each other. The result is that in non-Gaussian noise environment, the performance of the detector based on locally optimum is better than conventional square detector.Detection technique based on locally optimum structure is two-dimensional and the calculation is complex, so the signal received by the receiver should go through coordinate transformation to get two parameters: amplitude and phase. Then apply locally optimum detection to amplitude and the phase is simply delayed without any other processing, the original two-dimensional structure is changed to one-dimension, and time-domain amplitude signal processing technology (ADP) is obtained. Therefore, ADP is acquired from dimensionality of locally optimum detection. When there is more than one case of interference, it will present Gaussian distribution according to the Central Limit Theorem. While the amplitude signal processing technology is transparent to the Gaussian distribution, so ADP could not inhibit multiple interferences, which could be told apart in frequency domain. The amplitude processing techniques is transplanted to the frequency domain to obtain frequency-domain processing techniques (FADP). Apply FADP to spread spectrum navigation system to get a new signal detection structure. Simulation result shows that FADP is capable of suppressing non-Gaussian interference. By increasing the noise ratio of input signal, FADP could enable spread-spectrum navigation receiver detect weak signals normally in complex channels.