Research On 2D Interferometric Signal Processing

Based on the one-dimensional interferometry,two-dimensional interferometry is developed.Compared with one-dimensional interferometry,two-dimensional interferometry can eliminate the influence of part of the environmental noise and constraint the targets’relative position strongly.It can solve the problems of target position and attitude measurement,which is the current research hotspot.In order to solve the problems of target attitude measurement based on the same-beam interferometry,observation target on image position and attitude measurement in two-dimensional interferometry,the main contents of this paper are followed.Based on the traditional interferometric signal processing,a multi-baseline same-beam two-dimensional interferometric method is proposed for spacecraft attitude measurement firstly.A mathematical model between the spacecraft attitude and the "spacecraft transmitting antenna-station receiving antenna" differential distance is established and a Newton iterative method to obtain spacecraft attitude is proposed.Meantime,its accuracy is evaluated by numerical simulation.The results show that when the interference frequency is 2.2GHz and the precision of extracting the interferometric phase is 0.1257 rad,the solving precision of the spacecraft attitude can reach 0.01°.As two-dimensional interferometric processing is extended to the images,a method to process image signals,named as two-dimensional interferometric correlation,is proposed in this paper.The phase shift/offset between signals is obtained by phase filtering,unwrapping and offset calculation.A slide stacking method is proposed to suppress the interferometric phase noise,which is helpful to solve the phase shift/offset.The results show that the accuracy of phase shift is 0.01 rad between image signals with no offset noise.Finally,based on the two-dimensional image interferometry,the position measurement and attitude estimation of observation target on image are carried out.The integer offset between the same landmark of two images is obtained by the initial offset search(feature point matching),and the complete offset between the landmarks is obtained by combining the decimal offset obtained by interferometry firstly.And the position of the landmarks of the observation target on image is obtained by coordinate reconstruction.The mathematical model of observation target on image attitude transformation is established according to the corresponding geometric relationship of several groups of landmarks before and after observation target on image attitude transformation.By comparing the characteristics of solving attitude transformation with particle swarm optimization and Newton iteration,a method(an initial value obtained by particle swarm optimization,and then applied for Newton iteration to obtain the result)of solving observation target on image attitude transformation is proposed,and the feasibility of this method is verified by experiments.