Research On Several Key Problems Of Microwave Staring Correlated Imaging Based On Aerostat Platforms

Microwave staring correlated imaging(MSCI)is a novel staring imaging technique based on temporal-spatial stochastic radiation field,which is typically generated by multiple transmitters with independent stochastic transmitted signals.The high resolution imaging results can be obtained by the correlation process of the collected echo and the radiation field.To realize continuous gaze and imaging of a specific area,this thesis conducts research on microwave staring correlated imaging based on aerostat platforms,and mainly solves the following problems:(1)the real transmitted wave-forms are somewhat different from those with preset parameters;(2)the transmit and receive antennas are in translational motion with the aerostat platforms;(3)the amount of computation and storage required for imaging large target areas may not be met by the limited resources on the aerostat platforms.To solve the first problem,a correlated imaging model containing gain-phase and synchronization errors is derived,and an error calibration method based on an extra reference receiver is proposed.The reference receiver collects the direct signal from multiple transmitters while the radar observes the target area;thus,the calibration of the errors is divided into two main steps.The First step is to estimate the gain-phase and synchronization errors using the direct signal collected by the reference receiver.The second step is to compensate the imaging model with the estimates from the first step,and then obtain the imaging results via a correlated processing algorithm.Compared with the methods which jointly reconstruct the targets and estimate the errors only based on the observed data by the receiver,the proposed method is easier to implement and can achieve better performance.Besides,the influence of transmitted waveform parameters,along with the position and orientation of the reference receiving antenna,on error estimation is analyzed,and the criterion for waveform design and antenna placement is derived.Then,an imaging model based on an aerostat platform is derived to solve the second problem.Based on the model,this thesis analyzes the effects of the platform's translational motion and the velocity measurement error on the imaging quality.By converting the imaging problem containing velocity measurement error into a minimization problem,joint processing algorithms for estimating the velocity of translational motion and imaging are proposed based on the Newton's method and Particle Swarm Optimization.Furthermore,an adaptive imaging method based on narrow-pulse for targets appearing in discrete clusters is proposed.In order to avoid the huge coefficient matrix resulting from finely meshing the whole imaging area,the proposed method exploits the narrow-pulse stochastic signals and the matched filtering method robust against gridding error to adaptively locate the target areas,and only the areas of interest are discretized to a fine grid.Therefore,the complexity of the imaging process is significantly reduced.Finally,considering the demand of the ability to produce narrow pulses,and the fact that the last level circuit of the actual transmitted channel in a narrow pulse system is usually the power amplifier requiring high response speed,this thesis conducts research on a microwave high-power narrow pulse amplifier.To further improve the flexibility and adaptability of the system,a microwave high power variable pulse-width amplifier is designed.The design process mainly solves two problems including the high-speed drain pulse modulation circuit design and RF pulse envelope depression.