Research On Spot Position Detection Technology Of Space Laser Communication Based On Array Detector

In the past few decades,thanks to the tremendous progress in the performance of optoelectronic devices and the rapid growth of the market demand for wireless communication equipment,the space laser communication technology has developed rapidly.It is considered as one of the important solutions to achieve ultra-high-speed,large-capacity space communications.However,due to the small divergence angle and the long distance,the laser will be interfered by some factors such as the vibration of the satellite platform and the turbulence of the atmosphere,which may make the beam deviate from the best direction and even cause the communication cut.Therefore,the space laser communication system needs high-precision spot position detection technology to realize the Acquisition Pointing and Tracking of the laser beam.Compared with CCD(Charged Coupled Device)and PSD(Position Sensitive Detector),the array detector can provide sufficient detection field of view by expanding the number of elements while maintaining high resolution and frame rate,which can meet the performance requirements of coarse tracking stage and fine tracking stage.It is an ideal choice for single detector compound-axis tracking technology.This paper mainly researches the spot detection technology of space laser communication based on array detector.The main research contents are as follows:1.The application of array detector in space laser communication is investigated,and several common photoelectric position detectors are compared.Based on the analysis of the array detector,a simplified model of the spot position detection system is established and its performance evaluation parameters are analyzed.The radius range of the spot which can make the array detector output continuously is discussed.2.Two new spot detection algorithms are proposed for array detector in the fine tracking stage:(1)A BP(Backpropagation)neural network position detection algorithm is proposed for Gaussian spot.In this paper,BP neural network is used to fit the functional relationship between the calculated value and the actual position of the spot center.A neural network structure optimization method is proposed for the specific application of spot position detection.The simulation results show that the algorithm achieves nanometer level detection accuracy while maintaining a low amount of calculations.In addition,it has the ability to suppress non-uniformity.(2)An algorithm for detecting the position of L-G(Laguerre-Gaussian)spot is proposed.The response of the array detector to different modes of L-G spot is discussed in depth.Based on the infinite integration method,the error caused by the detector shape and the dead zone is compensated by introducing an error factor.The simulation results show that compared with the infinite integration method,the algorithm can reduce the root mean square error by 97.6% at most,and its position detection accuracy can reach um level.3.This paper provides a feasible solution for the detection of the spot position of the array detector in the coarse tracking stage.Firstly,in order to accurately calculate the current value of the basic unit of the multi-element array detector,an improved infinite integration method is proposed.On the basis of the original algorithm,the dead zone and the excess integration area outside the detector boundary are eliminated.At the same time,by introducing an error factor,the error caused by the inaccurate size of the dead zone is compensated.Then,taking a 4 × 4 multi-cell array detector as an example,three strategies are proposed for light spots of different sizes.Finally,the threshold is defined to determine the spot coverage of the basic unit accurately,so as to further determine the required strategy.4.The hardware circuits are designed for the four element array detector and the multi-element array detector respectively,and the displacement test platforms are built for them respectively.The proposed BP neural network position detection algorithm and multi-element array detector position detection solution are experimentally tested.The experimental results show that the spot detection system using the multi-element array detector position detection solution and the BP neural network position detection algorithm can meet the index requirements of the coarse tracking stage and fine tracking stage of the APT system,respectively.