Research On Knowledge-Assisted Constant False Alarm Target Detection Technology

Constant False Alarm Rate detection is the most common method to control the false alarm rate in radar automatic detection systems.The maritime environment is complex and changeable,and sea clutter interference has always been the main reason that affects the detection performance of radar CFAR.The CFAR detector is established based on the statistical distribution characteristics of clutter.The deviation of the parameter estimates of the CFAR detector will result in constant false alarm loss,and the greater the deviation,the greater the constant false alarm loss.Therefore,it is a reliable way to improve the radar detection performance to obtain the statistical distribution characteristics of real-time sea clutter reasonably and effectively,and use the characteristics of sea clutter to adjust the CFAR detector adaptively.In this thesis,a knowledge-assisted comprehensive CFAR detector is constructed,aiming to obtain stable false alarm rate and high target detection probability in complex sea clutter environment.The research content includes the following parts:(1)Under the background of K-distributed sea clutter,an algorithm for determining the parameters of the sea clutter distribution model based on radar parameters and environmental information is improved.On this basis,a CFAR detector based on the cognition of sea clutter distribution characteristics is proposed.The K distribution model can not only describe the amplitude distribution characteristics of sea clutter relatively accurately,but also reflect the correlation between sea clutter data.By constructing the relationship between the K-distributed sea clutter distribution parameters and the threshold factor of the CFAR detector,the environmental adaptability of the threshold factor of the CFAR detector is improved.The experimental results show that under the background of K-distributed sea clutter,the CFAR detector based on the cognition of sea clutter distribution characteristics proposed in this thesis has better detection ability and robustness than the CFAR detector with fixed parameters.(2)Build the Radar My SQL threshold factor knowledge base.Under the background of K-distributed sea clutter,the corresponding relationship between the CFAR detector threshold factor and K-distribution parameters obtained by numerical analysis is stored in the radar threshold factor knowledge base as prior knowledge,and different types of CFAR detectors with different false alarm probabilities are constructed.Threshold factor knowledge base under condition.The radar perceives the K distribution parameter information of sea clutter,and through the intelligent update scheduling system,finds,invokes and updates the corresponding threshold factor in the knowledge base,and realizes the adaptive control of the CFAR detector threshold factor.(3)Construction of a comprehensive CFAR detector system.According to the radar echo information and the perceived sea clutter environment information,not only the parameter control of the CFAR detector can be realized,but also the reference window structure of the CFAR detector can be adaptively adjusted.The radar echo information is used to obtain the size of the extended target in the radial direction of the radar,and a knowledge-aided comprehensive CFAR detector is proposed.Adaptability to uniform clutter.Experiments show that the integrated CFAR detector can obtain stable false alarm rate and high target detection probability in complex sea clutter environment.