Study on Virtual-Measure Kalman Filter Algorithm in Radar Networking

Using Kalman filter algorithm (KFA) in tracking target in radar networking system (RNS), measure-value of target in networked radar (NR) polar coordinate system has the nonlinear relation with state-value of target in fusion center rectangular coordinate system of RNS. The nonlinear relation does not satisfy linear requirement of KFA application. So this paper virtualizes fusion center rectangular coordinate system as the measure coordinate system of KFA. Through this way, original nonlinear relation is simplified as a linear form. By means of modeling noise of virtual measure coordinates, and constructing the initialization strategy, KFA can be used to solve the problem of state estimation in RNS. The simulating verification shows that virtual-measure KFA proposed in this paper is more precise than extended KFA (EKFA) used for state estimation in RNS.