Method for imaging actual aperture foresight on basis of incomplete data deconvolution

Abstract

The invention discloses a method for imaging an actual aperture foresight on basis of incomplete data deconvolution, which mainly solves the problem of imaging the right ahead scene of a flightpath. The method has the processing procedures that: 1) sequence overlap scanning is carried out in a ground monitoring area for obtaining the radar return data; 2) linear interpolation is utilized to expand the radar return data for obtaining the approximate complete deconvolution data; 3) the frequency-domain response of the approximate complete deconvolution data is calculated; 4) a frequency-domain Gaussian lowpass filter is defined; 5) the frequency-domain response of the approximate data is multiplied by the frequency-domain Gaussian lowpass filter, and the frequency-domain response of the processed return data is obtained; 6) the frequency-domain response of an antenna pattern is calculated; 7) a space-invariable filtering algorithm is adopted for calculating a transmission function; 8) the frequency-domain response of a deconvolution post restored image is calculated; and 9) a restored image is gained. The method has the advantage of improving the imaging quality and the detection speed, and can be used for an airborne radar monitoring system in the data processing field for imaging the ground scene of the flightpath.

Description

technical field [0001] The invention relates to the field of imaging, in particular to the problem of high-resolution imaging of the scene directly in front of the flight track, which can be used for high-resolution imaging of the scene along the direction of the flight track by monitoring systems such as airborne radar. Background technique [0002] In the ever-changing modern high-tech warfare environment in the future, battlefield information is complex and changeable, and fighters are fleeting. Timely and correct battlefield detection and tactical reconnaissance are related to the success or failure of the war. Therefore, it is necessary to require the radar imaging system to have a certain imaging accuracy and imaging range. However, when the antenna beam is close to coincident with the track line, since the ground targets distributed on both sides of the track have the same Doppler history, aliasing is likely to occur, and the Doppler change rate of the target is very ...

AI Technical Summary

Problems solved by technology

Although monopulse angle measurement has high measurement accuracy in theory, in practical applications, there are often the following limitations: when a complex-shaped target moves relative to the radar, it will cause the deviation between the apparent center of the target and the actual center of the target, Angular flicker phenomenon occurs, which restricts the imaging quality; the single-pulse forward-looking imaging technology cannot measure the angles of different scattering centers in the same range unit, and can only obtain the positions of equivalent scattering points, which reduces the accuracy of angle measurement; when the same beam memory When there are many targets, the angle measurement accuracy will drop sharply and even the position of the target cannot be detected accurately, especially when there are two or more targets with similar energy in the beam azimuth range within a certain distance unit, the angle measurement accuracy is especially Low; in monopulse forward-looking imaging, each angle-measuring coordinate plane usually uses two independent receiving branches, namely two branches in the azimuth plane and two branches in the elevation plane, and the system is complex

However, there are also certain problems in the deconvolution process: single-channel deconvolution is prone to produce ill-conditioned solutions, and the algorithm requires a high signal-to-noise ratio. When the signal-to-noise ratio is less than 30dB, the target resolution effect is poor; To reduce the influence of ill-conditioned effects, Berenstein et al. developed multi-channel deconvolution technology for information reconstruction of linear shift invariant systems, which can transform part of the deconvolution problem into a good state, but not all convolution sets Both meet the strong coprime condition. When the multi-channel cannot satisfy the strong coprime condition, the signal-to-noise ratio will be greatly reduced after deconvolution. Therefore, the use of multi-channel deconvolution technology also has great limitations.

The research on this technology at home and abroad is based on fully convolutional data, but in actual situations, radar echo data is only a part of all convolutional data, so direct deconvolution for real-aperture forward-looking imaging algorithm will not applicable in this case

Images