Research On Spatial Correlation And Motion Compensation Of Radiation Field Of Microwave Staring Correlated Imaging

Microwave staring correlated imaging(MSCI)is a novel staring imaging technique,which constructs temporal-spatial stochastic radiation field(TSSRF)by multiple transmitters emitting stochastic modulated signals.The scattering echo is received by the receiver after the stochastic radiation field is scattered by the target.The target image can be recovered by correlated imaging process(CIP)between the received echo and the pre-calculated TSSRF.The randomness and the accurate calculation of the radiation field are necessary for MSCI to reconstruct the target.Therefore,the thesis focuses on the aforementioned two aspects.Aiming at the randomness of the radiation field,we study the influence of the spatial correlation of the radiation field on MSCI and propose some methods to optimize the radiation field via spatial correlation minimization.For the aspect of accurate calculation of the radiation field,we mainly concenter on the calculation error of the radiation field caused by the quasi-stationary platform's motion.The work of this thesis includes:The first part of this thesis is the optimization of the radiation sources based on the spatial correlation of radiation fields.First,the influence of the spatial correlation of the radiation field on MSCI is studied.The results show that the radiation field with lower spatial correlation has better imaging performance no matter MF or sparse recovery algorithm in Compressed Sensing(CS)is used.Then we propose two optimization algorithms for radiation sources using frequency hopping signal and multi-frequency signal respectively.For radiation source using frequency hopping signal,we optimize the positions of the transmitting antennas and the frequency-hopping code by simulated annealing algorithm.For radiation source using multi-frequency signal,the amplitude of each frequency component in the multi-frequency signals serves as the design variable.Considering the power and the amplitude constraints in practical radar systems,we first form the waveform design as a constrained optimization problem.Then it is reformulated as a nonsmooth unconstrained optimization problem by the penalty function method.With the smoothing approximation,the reformulated problem is solved by alternating minimization.The second part of the thesis mainly establishes the imaging model for MSCI based on quasi-stationary carrier platform and studies the influence of the platfonm's motion on the imaging performance of MSCI.We decompose the complex motion of the antenna array into translations along the three coordinate axes and rotations around the three coordinate axes,and incorporate these motion parameters of the antenna array into the imaging model.Therefore,the established model is accurate for MSCI based on the quasi-stationary carrier.Then on the basis of this model,we study the influence of the carrier's translation and rotation in the forms of uniform velocity,uniform acceleration and sine function on MSCI through numerical simulations.The simulation results show that the carrier's motion will degrade the imaging performance of MSCI seriously.So further research is necessary for MSCI based on the quasi-stationary platform.The third part of the thesis focuses on MSCI based on the quasi-stationary platform which is assumed to translate and rotate uniformly.As MSCI has the advantage of fast imaging,it can be approximately assumed that the antenna array' s translation and rotation are uniform during a coherent imaging interval.On the basis of this assumption,the imaging equation whose radiation field matrix depends on the translational velocity and the rotational angular velocity of the antenna array is obtained.The measurements data of the sensors is used to make compensation for the carrier's motion at first.To eliminate the residual measurement errors after sensors'compensation,two self-calibration imaging algorithms based on the Newton's method and the particle swarm algorithm are proposed.The proposed algorithms can estimate the measurement errors while imaging and remove the influence of measurement errors on MSCI.The fourth part of the thesis studies the MSCI based on the quasi-stationary carrier platform whose motion is assumed to be random.First,the imaging model established in Chapter 3 is simplified.The movement of the antenna array during the transmission of pulses and the reception of echoes is ignored,and only the movement of the antenna array during the delay between transmission and reception and the interval between different pulses is considered.On the basis of the simplified model,the imaging equation for carrier platform with random motion is obtained,and the motion parameters of the antenna array are measured for preliminary compensation.Then a sparse self-calibrating imaging algorithm for measurement errors is proposed for sparse target.The sparse self-calibration imaging algorithm uses Bayesian learning as the framework,and iterates between target reconstruction and measurement errors estimation until convergence.The algorithm can decrease the residual measurement errors,and improve the imaging performance of MSCI based on quasi-stationary carrier platform.