Research On Atmospheric Noise And Narrowband Interference Processing In VLF Communication

Very low frequency(VLF)communication technology has important applications in military and civil fields.The communication performance of VLF is affected by atmospheric noise and narrowband interference.The purpose of this thesis is to study the atmospheric noise and narrowband interference suppression algorithm to reduce communication bit error rate.Recent researches show that the probability density function of atmospheric noise has obvious heavy-tail feature.In view of this,the distribution models of impulsive noise can be categorized into three types,i.e.,the symmetric ?-stable distribution,the Class A distribution,the Gaussian mixture model.In addition,according to the different characteristics of narrowband interference,it is modeled as single-frequency interference,multi-frequency interference and narrowband Gaussian noise.The impulse noise can not provide closed-form probability density functions in most cases.Therefore,the maximum likelihood detection and locally optimal detection need numerical calculation,which are difficult to apply.The traditional nonlinear transformation methods have some weaknesses such as suboptimal detection performance,fixed tail of nonlinear function,and can only suppress the impulse noise with specific distribution,etc.In this thesis,the power law nonlinearity function is constructed and the efficiency function is introduced as the design index.The simulation results present that the power law nonlinear design method can obtain the optimal detection performance for the three kinds of impulse noise with advantages of being easy to analyze and apply.In order to suppress narrow-band interference,this thesis introduces two suppression methods in transformation domain,i.e.,whitening filter and subspace filter.Both methods need to estimate the covariance matrix based on the received data,and perform eigendecomposition on it.The whitening filter uses reciprocal of eigenvalues for interference suppression while the subspace filter mainly identifies subspaces of signal,interference and noise by the eigenvalue of the eigenvector.The simulation results present that the subspace filter is more robust and efficient to suppress single-frequency interference and multi-frequency interference,while the whitening filter is more suitable to suppress narrow-band Gaussian noise.This thesis also considers the atmospheric noise and narrowband interference in the VLF channel,and combines the nonlinear processing method with the interference suppression technology to design a joint suppression scheme.The simulation results present that it is necessary to adjust the specific suppression schemes according to the different problems and scenes in the practical applications.In this thesis,the subspace power law trailing method is introduced to jointly suppress atmospheric noise and single-frequency interference,atmospheric noise and multi-frequency interference.The whitening power law trailing method is introduced to jointly suppress atmospheric noise and narrow-band Gaussian noise.At last,the main work and the innovation points of this thesis are summarized,and the possible research works are prospected.