Research On Radar Resource Allocation And Target Tracking Methods

With the rapid development of radar technology,radar functions are becoming more and more complex,and task modes are becoming more and more.How to improve radar detection capabilities and target tracking accuracy with limited resources has become an important research topic.According to the technology route from the allocation of radar resources to target tracking,from measurement noise to measurement equation,from linear filtering to nonlinear filtering,and from non-maneuvering tracking to maneuvering tracking,technical research is being conducted in the two technical areas of resource allocation and target tracking in this dissertation.The main work is as follows:1.Conduct research on radar resource allocation,including timing design and task scheduling algorithm.(1)Timing design: The definition of dynamic time slice is proposed,and the division of dynamic time slice is proposed according to different application scenarios of radar.The timing design is based on the dynamic time slice as the pace,which overcomes the shortcomings of the traditional scheduling interval fixed,inflexible,easy to generate time fragments,and low time utilization when scheduling tasks are less.In the proposed algorithm,the scheduling interval is flexible and variable,without time fragmentation and high time utilization.The effectiveness of the timing design is proved by computer simulation.(2)Task scheduling algorithm: An adaptive scheduling algorithm based on impact of tasks is proposed.For conventional task scheduling algorithms,the impact of tasks is not taken into consideration,that is,the impact of current scheduling tasks on subsequent tasks.In fact,the execution of current tasks has an important impact on subsequent tasks.This dissertation analyzes the correlation between current tasks and subsequent tasks.The model of task loss and the model of temporal margin are built with respect to the impact between tasks.In addition,the notion of temporal margin is proposed to reflect the effect of the current task on the following tasks.Based on the above,an adaptive scheduling algorithm based on the impact of tasks is proposed.Simulation experiments show that the proposed algorithm could greatly improve the performance of multifunction radar in aspects of the number of found targets and temporal margin in both underload and overload situations.2.By analyzing the influencing factors of radar angle measurement accuracy and range measurement accuracy,the distance error and angle error are related to the SNR.The measurement noise covariance matrix can be appropriately modified according to the SNR estimation results.Combined with the advanced non-linear filter method,the extended Kalman filter method based on SNR model and the Unscented Kalman filter method based on SNR model are proposed.In the traditional nonlinear filtering tracking algorithm,the angle measurement error and the distance measurement error take fixed constant values according to experience.In fact,the angle error and range error are variable values,which will change with the change of SNR.The algorithm proposed in this dissertation performs tracking filtering by continuously correcting the measurement noise covariance matrix.Compared with the traditional non-linear filtering algorithm,the tracking performance of the proposed algorithm is significantly improved,which is reflected in the higher target tracking accuracy and faster convergence speed.The effectiveness of the algorithm is verified by computer simulation.3.Wideband phased array radar can obtain high resolution range profile(HRRP).In this dissertation,this feature is used to obtain the pose of the target and apply it to target tracking.The pose of the target is estimated in the real time by HRRP,and then the pose is added to the target measurement equation.Because the relationship between the pose and the motion parameters of the targets is nonlinear,this dissertation combines advanced nonlinear filtering algorithms to achieve target tracking.(1)For non-maneuvering target tracking,the extended Kalman filter algorithm aided by the pose of target(Pose-EKF)and the Unscented Kalman filter algorithm aided by the pose of target(Pose-UKF)are proposed.In the Pose-EKF algorithm,since the pose in the measurement equation is a non-linear function,the expression of the pose from non-linear to linear equation is derived in this dissertation.The results of simulation demonstrate that: compared with the traditional extended Kalman filter algorithm(EKF)and the traditional Unscented Kalman filter algorithm(UKF),the proposed algorithm can greatly improve the target tracking accuracy(position precision and velocity precision)and the convergence speed.The pose measurement error has little effect on the tracking performance.The difference of the tracking accuracy between Pose-EKF and Pose-UKF is little.But the Pose-EKF is better than Pose-UKF in terms of computation time,but Pose-EKF fails and Pose-UKF is effective when the pose is critical.(2)A maneuvering target tracking algorithm aided by the pose is proposed.Compared with the traditional algorithm,the proposed algorithm could greatly improve the target tracking accuracy.It has smaller target position error,speed error,and higher target prediction accuracy.At the same time,the influence of pose error on the performance of this algorithm is analyzed.The smaller pose measurement error is,the better the target tracking performance is.The effectiveness of the algorithm is verified by computer simulation.4.Research on target tracking algorithms in clutter environment,including non-maneuvering target tracking and maneuvering target tracking in clutter environment.(1)For non-maneuvering target tracking in clutter environment,based on the radial velocity measured by moving targets detection(MTD),Radial velocity track gate was established using radial velocity.Radial velocity track gate was deduced,radial velocity dimension was introduced to the measurement equation,and radial velocity of target observed was updated with radial velocity measured by MTD.Compared with the traditional filtering algorithm in clutter environment,the target tracking performance of the proposed algorithm is greatly improved.(2)Aiming at the problem of maneuvering target tracking in clutter environment,the radial velocity is obtained by the Doppler measurement,and the radial velocity dimension is added to the measurement equation of the target and the radial velocity in the measurement equation is linearized by carrying out Taylor series expansion and omitting high-order quantities.The radial velocity gate is added to the radar targets association to filter out more clutter points.The radial velocity of target observed is updated with the radial velocity calculated by Doppler measurement.Because more measurement information is used in the maneuvering target tracking algorithm,the target tracking performance is greatly improved compared with the traditional maneuvering target tracking algorithms in clutter environment,which is reflected in the target position accuracy,speed accuracy,and the accelerated convergence speed.When the target is maneuvering,the response speed of the proposed algorithm is faster.The influence of Doppler measurement error on target tracking performance is also analyzed.The smaller Doppler measurement error is,the better the target tracking performance is.The effectiveness of the algorithm is verified by computer simulation.