Research On Vessel Target Detection Method Under Sea Clutter Background In High Frequency Surface Wave Radar

In the application of target detection in high frequency surface wave radar(HFSWR),it is inevitable to be affected by many kinds of clutter,interference and noise.Amough them,the sea clutter is very strong and often higher than the target echo.On the one hand,it seriously interferes the ship target detection near the sea clutter area.While on the other hand,when the Doppler frequency deviation caused by the ship echo overlaps with the broadened sea clutter spectrum,the sea clutter will completely submerge the target and form a blind detection area.For the compact HFSWR system,due to array aperture reduction and antenna miniaturization,the radar beam width increases and gain decreases.The masking effect caused by sea clutter Doppler expansion becomes more serious,which increases the difficulty of target detection under sea clutter background.Therefore,to solve the problem of target detection under sea clutter,especially target detection within the broadened sea clutter area,is a key problem to be solved in target detection of HFSWR,and is also an important research direction of sea surface surveillance and monitoring by HFSWR.In this dissertation,the comparative analysis of sea clutter and target characteristics,sea clutter suppression and target detection in sea clutter are studied.The main contents are as follows:Firstly,based on the generation mechanism of sea clutter and its modulation effect on echo,the characteristics of sea clutter and target in radar echo are analyzed.To study the particularity of target detection in the background of HFSWR sea clutter,the multi-domain characteristics of sea clutter and target in time domain,Doppler domain,space domain and joint domain are compared and analyzed.In particular,to reveal the difference between sea clutter and target in joint domain,the measured data of large array narrow-beam HFSWR is introduced to carry out analysis.And the comparison analysis is carried out with small array broad-beam HFSWR of poor spatial resolution.It is shown that the main energy of sea clutter is concentrated in a small number of directions.When the narrow beam matches the sea clutter to its coming azimuth,the energy of sea clutter in each single directions is significantly lower.The influence factors of radar accumulation time,relation between target and sea clutter area and shipborne platform are analyzed in combination with the situation of target near and within sea clutter area,which provides an important basis for the following researches on sea clutter suppression,sea clutter information extraction and target detection within sea clutter.Secondly,to solve the problems existing in sea clutter suppression,the traditional time-domain cancellation method is improved,the matrix decomposition and time-frequency combination method for sea clutter suppression is proposed,and the space-time-frequency multi-domain method for sea clutter information extraction is developed.Since the traditional time-domain sea clutter cancellation method is difficult to distinguish the target from the sea clutter,which leads to the problem that the target is wrongly eliminated during the cancellation,a boundary-constrained time domain cyclic cancellation method is developed.Based on the sinusoidal modulation of sea clutter in time domain and the boundary statistical properties in frequency domain,the method model and estimate the parameters of the echo signals in time domain,and the cancellation process is constrained.Furthermore,a combined matrix decomposition and time-frequency suppression method is proposed based on the time-frequency domain characteristics of sea clutter.Combined with the instantaneous frequency estimation of the echo,the sea clutter suppression is realized from the perspective of matrix eigenvalue decomposition and time-frequency analysis,which overcomes the problem that the singular value of the former is not easy to be determined,and avoids the mutual restriction between the time and frequency resolution of the latter.It further improves the suppression ability of the sea clutter.However,the target that is always reside in the sea clutter area during the accumulation period,cannot be detected.Therefore,combined with space,time and frequency domains,a sea clutter information extraction method is proposed,which improves the extraction accuracy of sea clutter information and lays a foundation for the subsequent target detection in sea clutter.Thirdly,to solve the problem of blind detection area caused by sea clutter,oblique projection operator is introduced into sea clutter suppression,the improvement of oblique projection algorithm and the construction of multilevel spatial filter are studied,and the target detection within sea clutter is realized under the three-dimensional detection of joint azimuth-range-Doppler domains.Considering the amplification effect of traditional oblique projection spatial filter on noise power and the reduce of output signal to noise ratio caused by small principle angle,the clutter suppression cost parameter is introduced.Therefore,an improved oblique projection spatial filter is proposed,and its special properties are derived.Through theoretical and simulation analysis,it is shown that the improved oblique projection filter inherits and develops the advantages of oblique projection,especially when the noise ratio is known,the signal to clutter plus noise ratio is maximized,which is more suitable for the construction of spatial filter for sea clutter.However,the oblique projection spatial filter is a single notch filter.Considering the multiazimuth characteristic of the sea clutter,the wide-beam internal sea clutter filtering is realized through the construction of multilevel spatial filter based on the influence of parameter selection analysis in constructing the multilevel oblique projection filter.Combined with Doppler and range domains,a three-dimensional joint detection method based on oblique projection/improved oblique projection is proposed.Simulation and measured data show that the proposed oblique projection based and improved oblique projection three-dimensional joint detection method are more effective than the conventional beamforming based three-dimensional joint detection method,and Joint domain localized method.In the senerio of low signal to clutter ratio,the performance of improved oblique projection based method is better than oblique projection based method.Finally,in view of information source,related external information will be helpful for target detection in the background of sea clutter.The main reason for the difficulty of target detection in the sea clutter spectrum can be attributed to the lack of necessary prior reference information when the target echo and sea clutter coincide.It is difficult to obtain the isolated form of target echo and sea clutter.To solve this problem,the radar sea clutter spectrum is first constructed by using the sea state information,and the characteristics of sea clutter spectrum under different sea states are analyzed.Then,by adding target into the constructed sea clutter spectrum and using measured radar data,sensitive features such as spectrum width,peak frequency and Bragg peak ratio for target detection are extracted.Finally,by analyzing the influence of sea conditions and sea state measurement accuracy on sea clutter sensitivity characteristics,the anomaly detection threshold of relevant sensitive characteristics is determined.Herein,a target detection method in sea clutter based on sea state information is proposed.This method is not constrained by the characteristic difference between the sea clutter and the target required by the sea clutter suppression methods in previous domain used,and it avoids the potential adverse effect of the suppression algorithm on the subsequent parameter estimation of radar,which opens up a new way on the target detection in the sea clutter of HFSWR.The research results of this dissertation are expected to deepen the understanding of sea clutter and target characteristics of HFSWR,improve the performance of sea clutter suppression method,and promote the technology development of target detection in sea clutter,which has promising research value and application potential.