Research On Detection And Inversion Of Ionosphere Clutter In Sky-surface Hybrid Radar

The High-Frequency Sky-Surface Wave Hybrid Radar is a new type of HighFrequency Over-the-horizon Radar,which combines the characteristics of HighFrequency Sky-wave Radar and High-Frequency Ground-wave Radar.It uses the principle of “lunching to sky,receiving from surface” in detection to achieve the region monitoring and target detection at long-range.It also has some anti-stealth capabilities.Thus it has broad prospects for development both in the fields of scientific research and national defense.However,its clutter background of spectrum more complex.There by,sea clutter and ionospheric clutter pollution are serious,which will affect the radar detection performance.Meanwhile,the ionospheric clutter can also be regarded as a kind of “echo signal” of the ionosphere detection under its generation mechanism.And the ionosphere clutter do have some similar characteristic to the which of targets,so it can be considered that the clutter also contains the information of the ionospheric abnormal states.The study of the detection and inversion methods of ionosphere clutter in radar echo spectrum will not only conduct an approach of detection of abnormal states in ionosphere,by which the field of radar's application can be expanded,but also provide some prior knowledge of the suppression of ionosphere clutter.This paper starts with the signal model of the sky-surface wave hybrid radar.Firstly,the radar signal path from the parts of sky wave and ground wave is analyzed.Among them,through the study of ionosphere structure characteristics and electromagnetic wave transmission characteristics,the calculation method of sky-wave and surface-wave path based on the spherical earth theory is given.Based on the assumptions of ionosphere clutter generation mechanism,the corresponding clutter signal path models are given.Then,from the viewpoint of the electric field,the spectral model of the signal received with the FMCW signal is deduced.In order to emerge the influence of the ionosphere on the mixed-path signal,the ionospheric reflection coefficient is introduced.Based on the electromagnetic field theory,the electromagnetic properties of the anomalous ionosphere structure(the source of ionosphere clutter)are analyzed,and the expressions of the equivalent scattering area and reflection coefficient are given.Combining the clutter signal path model and the spectrum derivation of the received signal,the clutter spectrum model is given.Finally,according to the characteristics of ionospheric clutter,this paper studies a method of detecting ionospheric clutter based on the related theories and tools of machine learning and neural networks.In this paper,the signal RD(RangeDoppler)spectrum is regarded as images,and a large number of simulated RD spectrum images are generated by the use of signal model and clutter model.These samples will complete the migration learning of the deep convolutional neural network to obtain an ionosphere clutter detector.Then,based on the results of the clutter detection,an inversion method for ionospheric clutter is proposed.Using the simulation results of clutter model,the correspondence between clutter characteristics and ionosphere anomaly parameters is established in multiple dimensions,in which the clutter detection results can be directly transferred to the ionosphere abnormal state parameters.Experiments through real RD data have proved the feasibility of the ionospheric clutter detection and inversion method proposed above.