Research On PAM Signal Correction Algorithm For Free Space Optical Communication

The constant innovation of science and technology has brought about the constant change of The Times,The human society has made great strides from the farming times,the industrial age,the information age,the data age and the artificial intelligence age.The demand for communication quality is also getting higher and higher.In the past,the era of electric communication has been replaced by the era of optical communication.Portable wireless communication and high-speed optical fiber communication have been used in every household.However,they all have their advantages and disadvantages.I introducing the new type of space optical communication technology,it combines the advantages of wireless communication with optical fiber communication and can play a huge role in solving the "The last mile"scenario.However,he needs to overcome the unavoidable problem of atmospheric channels affecting the quality of communication.This article is based on this expansion,The machine learning algorithm in artificial intelligence is used to correct the signal received by the receiving end.In this paper,the basic theory of free space optical communication is introduced firstly.Including the basic models describing atmospheric turbulence,and the model of light intensity undulation and phase undulation.And also introduced several parameters and concepts described the atmospheric turbulence.Then we introduce the atmospheric phase screen which is used to simulate atmospheric turbulence and simulate and compare it by two methods.Based on one of the methods,the superposition of multi-phase screen simulates the transmission of laser light source in a turbulent atmosphere,then with Taylor frozen turbulence hypothesis,I calculate the intensity of light scintillation and phase fluctuations with time.Finally,based on the model of light intensity scintillation,first using the OOK signal modulation for simulation calculation.And then PAM method is used to simulate the PAM signal changes with time.After the time is divided into blocks,the machine learning method is used to classify the decision by the support vector machine,combined with blind detection and optimal theory detection,analyze and compare the signal correction function of SVM.