Research Ontarget Tracking Algorithm For Radar Network

The radar network consists of multiple widely distributed radar nodes,and it can be used to detect and track the targets effectively.As the radar network can observe the targets via different angles of view and obtain sufficient measurements of the targets,it generally becomes a focus in the radar research field.The traditional radar network usually handles the detector and the trackor separately.The fixed detection parameters and the measurements with different amount of uncertainty make a loss of effective measurement information and cause the redundancy of measurement information,which decrease the detection performance of the radar network.Target tracking is one of the most important functions of the radar network.Due to the more careful processing of the measurement information and better selection of the target observation views,the spatial diversity gain can be effectively improved,and the efficiency of measurement data can also be improved,which leads to an improvement of the tracking performance.In this dissertation,the joint parameter adjustment and the measurement data selection for networked radar target tracking are researched through the theoretically study,method analysis,simulation experiment verification,and so on.The main contributions of the dissertation are listed as follows:1.A novel radar network tracking method based on jointly adjustment of the detecting and tracking parameters is proposed.By constructing a closed feedback loop between the detector and the trackor,the adaptive parameter adjustment is achieved.At the same time,the detection threshold and the measurement data passby the threshold can be dynamic adjusted.It can be used to avoid the performance loss caused by separate consideration of the detector and trackor which is very common in the tradictional radar system.In this way,the tracking performance of the radar network is improved.2.A measurement selection strategy and target tracking method based on sparsity matrix decomposition for the synchronous radar network are proposed.Through the sparsity decomposition of the covariance matrix of the measurement data,an efficient measurement selection for target tracking is achieved.The complex signal processing for the non-selected measurements with low-confidence is avoided.The computational complexity of the radar network is reduced.The timeliness and the performance of target tracking for radar network are improved.3.A measurement selection strategy and target tracking method based on measurement information compensation for the synchronous radar network are proposed.By jointly considering the measurement information and the measurement compensation between different measurements,the measurement values are effectively screened and the target tracking performance of the radar network is improved.4.A measurement selection strategy and target tracking method based on measurement information compensation for the asynchronous radar network are proposed.Considering the different scanning periods for the radars in the asynchronous radar network,the measurement data with high value are effectively selected and the tracking performance of the asynchronous radar network is improved.The proposals in this dissertation are verified by numerical simulations.Simulation results show that by means of the proposed target tracking methods,the target tracking performances in the radar network can be efficiently improved.