Research On The Modeling And Processing Of Atmospheric Noise

Submarine is one of the important weapons in modern war and it is drawing more and more attention, while the reliable command and communication of the submarine base is always a problem. Long wave is the electromagnetic wave whose frequency is between 3Hz and 30 kHz, and it is one of the most important method of marine communication. The main factor that affects the performance of the long wave communication receiver is the interference of short-time pulse noise, which has the characteristics of short duration and high instantaneous amplitude. For the purpose of mitigating the influence of short-time pulse noise, this thesis analyzes the main mechanism of pulse noise generated in the long wave communication, and a physical sferic noise model based on Poisson cluster distribution is established, and then the correctness of the model is tested. This thesis also looks for the signal processing method to suppress the impulse noise and improve the performance of the receiver. And then some commonly used nonlinear filters are introduced, and we compare the performance of different nonlinear filters in MSK coherent receiver. At last the factors that affects the performance of receiver and the optimal threshold of hard limiter are analyzed.In the beginning of this thesis, the main features of long wave communication are introduced. And then the main models of atmospheric radio noise and the methods which are used to suppress pulse noise are provided.The common models of atmospheric radio noise are discussed in second chapter of this thesis and a statistical physical model which is according to the physical mechanism of atmospheric radio noise and making use of the theory of electromagnetic wave propagation and finally based on that impulse noise arrival times follow Poisson cluster distribution is established. Several factors which impact the amplitude of atmospheric radio noise are analyzed and some appropriate amendments are made in this model.The statistical physical model realization and parameter settings in matlab are discussed in third chapter of this thesis. First of all, this model is realized in matlab. And then some simulation results are analyzed which verify the correctness of model. And finally, a conclusion that ?-stable distribution model can well describe the atmospheric radio noise is made.MSK receiving and several commonly used nonlinear processors are introduced in fourth chapter of this thesis. And then the simulation results showed that the performance of different nonlinear processors are different under AWGN and atmospheric radio noise. After that, several factors that affect the receiver performance and the best threshold of hard limiter are analyzed.