Research On Multi-mode Superimposed Vortex Beam Generation And Transmission Optimization Method

In the age of rapidly developing communication technology nowadays,the channel capacity and information rate requirements of optical communication systems have become higher as the demand for network traffic and bandwidth grows.An effective way to achieve highcapacity communication is mode multiplexing technology,which loads information in multiple channels to increase the system channel capacity.By virtue of different topological charge vortex beams(VBs)being orthogonal,Orbital Angular Momentum(OAM)can be used as a mode multiplexing channel to meet the requirements for large channel capacity,and it has received wide attention and deep research.Compared with single-mode VBs,multi-mode superimposed VBs can coaxially transmit multiple OAM modes that are separated from each other,avoiding the complex problem of combining multiple OAM beams,and providing large mode multiplexing channels for optical communication systems.However,it is more complicated in generation and transmission,and thus it requires a suitable approach.Based on the study of the fundamental theory and key technologies in the field of multi-mode superimposed VBs generation and transmission,this thesis focuses on a multi-mode superimposed VBs generation scheme based on adapted mutation particle swarm optimization(AM-PSO)algorithm,a generation scheme based on neural network technology,and a transmission performance optimization scheme based on power pre-compensation,and the main work and innovations of the thesis are as follows.(1)A multi-mode superimposed VBs generation scheme based on AM-PSO is proposed to solve the problem that the traditional iterative algorithm generates VBs with large differences when the number of OAM modes is increased.Compared with the existing Pattern Search Assisted Iterative algorithm and Particle Swarm Optimization algorithm,simulation results show that the scheme can generate phase-only grating with lower Relative Root Mean Squared Error(R-RMSE)between the grating and the expected one,and higher energy conversion efficiency.The generated OAM modes are more consistent with the expected power distribution.The R-RMSE is 3.722×10-3 and the energy conversion efficiency is 92.97%when generating 20 evenly spaced OAM modes with equal power.AMPSO effectively improves the mode purity of the vortex beam.(2)A neural network-based multi-mode superimposed vortex beam generation scheme is proposed to construct a learning model of the decomposition coefficients of different topological charge,which solves the problem of the high complexity of the iterative method and achieves low R-RMSE generation of arbitrary topological charge combined 20 OAM modes.Simulation results show that the trained model has a good fit in the topological charge range of[-30,30].The loss is less than 3×10-4 during training and testing.The network has the ability to predict the mode decomposition coefficients with lower R-RMSE,which is 2.448×10-3.The energy conversion efficiency of the generated grating is 96.56%.(3)A multi-mode superimposed VBs transmission optimization scheme based on OAM mode power pre-compensation is proposed to recover the orbital angular momentum spectrum spreading effect caused by atmospheric turbulence and to solve the problem of inter-mode crosstalk occurring between different OAM modes in turbulent channels.The simulation results show that the scheme can effectively recover the OAM spectrum under atmospheric turbulence,improve the quality of the received VBs at the receiver,and enhance the communication performance of the free-space optical communication system in the case of 500m freespace propagation with the number of OAM modes of 10 and 20,and atmospheric turbulence intensity of 5×10-15 and 1×10-15,respectively.