Research On Resource Management Algorithm For Networked Radar Based On Target Tracking

The networked radar system can effectively enhance the target sensing performance and the system survivability by the networking and fusion of multiple radars that are widely distributed in space,which has been widely concerned and studied by scholars.Compared with traditional monostatic radar system,the networked radar system has higher flexibility and degree of freedom in resource allocation,and the closed-loop optimal allocation of system resources can be realized by resource management technology to fully exploit the potential of the networked radar system in target sensing.The core of resource management technology of networked radar system based on target tracking task is to design the corresponding closed-loop resource allocation algorithms according to diverse radar systems and different sensing scenarios.Therefore,based on the multi-target tracking task,this dissertation studies and designs diverse radar network resource management algorithms for different sensing scenarios and various radar network systems.The main work and contributions of this dissertation can be summarized as follows:1.The problem of joint node selection and multi-dimensional resource allocation in the radar network with separated transmitters and receivers is studied.The posterior cramér-rao lower bound(PCRLB),which contains optimization variables of transmitter selection,receiver selection,dwell allocation,power allocation and bandwidth allocation,is derived to characterize the tracking performance.Considering the system resource constraints,the optimization model of joint node selection and multi-dimensional resource allocation is established,with the sum of the PCRLBs being the objective function.For the non-convexity of the established optimization problem,a two-stage solution algorithm based on the sequential convex programming(SCP)and the semi-definite programming(SDP)is proposed.The closed-loop optimal allocation of the resource of the radar network is realized,which significantly improves the overall multi-target tracking performance.2.For the radar network composed of multiple phased array radar nodes,the problem of joint node selection and power allocation based on multi-target tracking task is studied.Considering the spectrum coverage between phased array radar network and communication electronic systems,the deviations between tracking accuracies and desired accuracies of multiple targets and the interference energy from radar network to communication electronic systems are adopted as objective functions,and the bi-objective optimization model is established.A hybrid particle swarm optimization(PSO)algorithm based on the Kullback-Leibler divergence(KLD)is proposed.The multi-target tracking performance of the phased array radar network is guaranteed,and the interference energy from phased array radar network to communication electronic systems is effectively reduced.3.For the radar network composed of multiple colocated multiple-input multipleoutput(MIMO)radar nodes,the problem of joint transmit beampattern design and resource allocation based on maneuvering targets tracking is studied.An optimization model with the low probability of intercept(LPI)performance constraints is constructed,which takes the sum of PCRLBs of maneuvering targets as the optimization metric.A two-stage optimization algorithm incorporating successive convex approximation(SCA)and cyclic minimization algorithm(CMA)is proposed to solve the optimization model.Under the premise of ensuring the LPI performance of colocated MIMO radar network,the overall tracking performance of multiple maneuvering targets is effectively improved.4.For the radar network of MIMO radars with orthogonal waveforms,the problem of joint transmit waveform selection and multi-dimensional resource allocation for multiple maneuvering targets tracking is studied.A multi-objective optimization model is established,with the maneuvering targets tracking performance and the radar network resource consumption being objective functions.The solution method based on the modified particle swarm optimization(MPSO)algorithm is proposed.The tracking performance of multiple maneuvering targets and the resource consumption of colocated MIMO radar network are both taken into account,and good compromise between them is achieved.5.In the presence of false alarm measurements,the problem of joint detection threshold adjustment and transmit resource allocation for colocated MIMO radar network is studied.The PCRLB containing detection threshold adjustment variable and information reduction factor(IRF)is derived in clutter.Considering maximizing multiple maneuvering targets tracking performance and minimizing radar network resource consumption simultaneously,a bi-objective optimization model is established.The simulated annealing based hybrid particle swarm optimization(SAHPSO)algorithm is proposed,which realizes good compromise between the optimization of maneuvering targets tracking performance and the optimization of colocated MIMO radar network resource consumption.