Parameter Estimation Of Phase Coded Signal In Impulsive Noise

As a typical signal with good anti-interference ability and low intercept probability characteristics,the phase-coded signal has attracted the attention of many scholars and has been widely used in military radar countermeasures and many civil communications.The Estimation of the carrier frequency and symbol rate of the phase-coded signal correctly and efficiently helps to obtain the modulation information of the recived signal,and thus occupy the communication control right.Therefore,it has become a research hotspot in communication signal parameter estimation.The traditional phase-encoded signal parameter estimation method schemes mostly use Gaussian distribution as the noise model.However,there are often a large number of non-Gaussian signals and noises in practical applications.This type of noise does not obey a Gaussian distribution and tends to exhibit significant impulse characteristics.The Alpha stable distribution is a more appropriate model for this type of impulse noise and is consistent with many actual datas.This type of impulse noise will result in a significant degradation or even complete failure of the signal processing method based on the Gaussian assumption.Based on this,the thesis studies the phase-coded signal parameter estimation under the alpha-stable distributed noise environment.The major work is outlined as follows:1.A new method based on cyclic correntropy function(CCF)for signal symbol rate estimation is proposed.The new method is simple to implement and does not require a priori information of noise.The simulation results show that the proposed method based on cyclic correntropy can effectively suppress impulse noise,especially in the environment of strong impulse noise,it has more obvious advantages than other algorithms.2.The signal processing method based on fractional lower-order statistics is a classical method for suppressing alpha stable noise,but it needs a priori information of the noise.The noise suppression performace depends on the selection of its order,which lacks theoretical support so far.In this regard,a new method for parameter estimation of phase-coded signals based on nonlinear transformation function is proposed,which is a cyclic autocorrelation method based on hyperbolic tangent(tanh)function.This method does not require a priori information of noise,and has a stronger ability to limit alpha stable distribution of noise outliers.The simulation results show that the proposed parameter estimation method based on nonlinear transformation has a more robust noise suppression capability,compared with the fractional lower order method.3.As a kind of Gaussian type function,the correntropy is complex in form and computationally intensive,and is not suitable for real-time calculation.In this regard,a new method based on quadratic rational kernel(RQK)function for signal symbol rate estimation is proposed.The method suppresses the impulse noise by constructing the quadratic rational kernel function of the signal,and then the symbol rate estimation under alpha stable distributed noise is realized.The proposed method does not require a priori information of noise,and the quadratic rational kernel function is simple in construction and small in computation,compared with the correntropy.The simulation results show that the proposed new method based on quadratic rational kernel function can effectively estimate the symbol rate of the phase-coded signal under both Gaussian noise and alpha-stable noise,which shows good robustness.And the parameter estimation performance is better than the fractional lower-order statistic under strong impulse noise environment.