Ionospheric Clutter Suppression Techniques Based On Blind Sources Separation For High Frequency Surface Wave Radar

High-frequency surface wave radar plays an important role in the coastal defense system by monitoring the exclusive economic zone within 200 nautical miles over-the-horizon in all-weather and all-time at low cost.However,the ionospheric clutter reflected by the ionosphere and received by the radar is strong and complex,which affects target detection.Ionospheric clutter mitigation is the key technology in high-frequency surface wave radar.Clutter suppression algorithm such as adaptive spatial processing has a decent performance on the clutter mitigation.But in real engineering applications of ionospheic clutter suppression,some non-ideal factors such as hetetogeneous clutter,main-lobe clutter and sheet clutter lead to performance degradation.In order to solve the above three key issues,this paper uses blind source separation technology to study the ionospheric clutter suppression under non-ideal factors.The main work of the full text is as follows:1.The ionospheric clutter characteristics are studied based on real data.Particularly,the time-domain independent characteristics and range-domain correlation characteristics of ionospheric clutter are analyzed which lay the foundation for the design of the clutter suppression algorithm based on blind source separation in the thesis.The structure of the ionosphere and the propagation mechanism of high-frequency radio waves in the ionosphere are described.The characteristics of different ionospheric clutter types are confirmed by the real data.In particular,the mutual independence of different electron density irregularities is analyzed,the correlation of the echoes from irregularities in range domain is studied.The research on the ionospheric clutter characteristics lays the foundation for the design of the blind sources separation algorithm.2.Regarding the problem of clutter covariance matrix estimation error faced by traditional adaptive spatial processing in heterogeneous ionospheric clutter suppression,a study on the reconstruction algorithm of ionospheric clutter covariance matrix based on blind source separation algorithm is carried out.The influence of heterogeneous clutter samples on the estimation of clutter covariance matrix is theoretically analyzed.The covariance matrix reconstruction algorithm based on blind source separation is studied: first,the clutter samples are separated via blind source separation algorithm,then the ionospheric clutter sources in the separated components are identified,and then the clutter power of Doppler cell under test is estimated using the ionospheric clutter source spectrum,finally the clutter covariance matrix is reconstructed by the estimated clutter power and ionospheric clutter wavefront.The covariance matrix estimation algorithm based on blind source separation breaks through the traditional SMI estimation algorithm that treats ionospheric clutter as a whole.The mechanism of heterogeneity is obtained by analyzing the internal structure of ionospheric clutter,which provides a new solution for ionospheric clutter suppression.3.In view of the performance degradation when the expected target direction of traditional adaptive spatial processing algorithm is close to the main-lobe ionospheric clutter,a blind source separation algorithm independent of expected target direction is studied.Meanwhile,the main-lobe clutter source identification after blind source separation under the framework of adaptive spatial processing is proposed to avoid performance degradation when the receiving beam direction of traditional adaptive spatial processing algorithm is close to the main-lobe ionospheric clutter.Directly using blind source separation on the received data including the detection unit can separate the target and the ionospheric clutter.Then target identification can help to find the target component and suppress the ionospheric clutter simulteneously.The direct blind source separation algorithm is unaffected by the expected target steering vector,but the separation performance is affected by the independence of the target and the ionospheric clutter.The influence of this non-ideal independence is theoretically analyzed.In addition,in the framework of the adaptive spatial processing algorithm based on blind source separation,a linearly constrained adaptive spatial processing algorithm based on blind source separation is proposed.The basic idea is to identify whether each clutter source is the main-lobe clutter.Then the main-lobe clutter wavefront is used as the linear constraint for adaptive spatial processing,and the sidelobe clutter sources are exploited for the clutter variance matrix estimation.The algorithm is still affected by the mismatch between the receiving beam direction and the real target direction,but it avoids the rapid deterioration of performance caused by the receiving beam direction and the clutter direction being too close.In addition,it is not affected by the non-ideal independence of the target and the ionospheric clutter.This algorithm and the direct blind source separation algorithm can suppress the main-lobe ionospheric clutter through cascade processing.Compared with the main-lobe clutter suppression method based on the eigen-decomposition blocking matrix preprocessing,the method based on blind source separation can be more accurate in estimating the wavefront of the main-lobe clutter source.It provides a new idea for the main-lobe clutter suppression problem.4.Refering to the problem of insufficient degree of freedom when the number of clutter sources is more than the number of antennas under the background of sheet ionospheric clutter,the blind source separation and adaptive processing algorithms is extended from the one-dimensional spatial processing to the two-dimensional range-beam joint domain algorithm to increase the number of degrees of freedom in the system.Based on the range-domain correlation characteristics of ionospheric clutter,the one-dimensional direct blind source separation algorithm is extended to the range-beam domain two-dimensional direct blind source separation algorithm.The dimension-reduced processing algorithm is studied and the maximum separable number of clutter sources is analyzed.The linear constrained adaptive spatial processing algorithm based on blind source separation is extended to the linear constrained fast-time domain STAP algorithm based on blind source separation,and the cascade processing of the two extended algorithms is studied.Range-beam domain processing can improve the algorithm's ability on multiple clutter sources at the same time,and provides an alternative method for the tough sheet ionospheric clutter suppression.5.In terms of the engineering application problem that a single algorithm is difficult to cope with ionospheric clutter suppression requirements under complex environment,the algorithm based on blind source separation studied in this paper is incorporated into the traditional ionospheric clutter suppression algorithm.Different clutter suppression algorithm is adopted on the corresponding ionospheric clutter region to achieve the optimal matching of ionospheric clutter characteristics and algorithms.The traditional ionospheric clutter suppression algorithm and the algorithm proposed in this paper are divided into one-dimensional spatial domain processing and two-dimensional range-beam joint domain processing,which are suitable for strip and sheet ionospheric clutter respectively.For each type of clutter,the relationship between the number of clutter sources and the number of freedom is further decided.Traditional adaptive processing algorithm is used for ionospheric clutter without sufficient degrees of freedom,and the cascade algorithm of direct blind source separation and linear constraint adaptive processing based on blind source separation is used for ionospheric clutter with sufficient degrees of freedom.The comprehensive ionospheric clutter suppression process gives a set of diversified and precise clutter suppression strategies based on clutter region division and different suppression algorithms for different regions,avoiding the limitations of a single ionospheric clutter suppression algorithm in complex ionospheric clutter environment.The ionospheric clutter suppression methods based on blind source separation proposed in this paper alleviate the three key problems of heterogeneous clutter samples,main-lobe clutter and sheet ionospheric clutter.The proposed methods enrich the theory of adaptive algorithms and expand the application scope of adaptive processing.A comprehensive suppression process for engineering application is proposed combined the new algorithms with traditional ionospheric clutter suppression algorithm to improve the overall performance of ionospheric clutter suppression.