An air-to-air missile vector scoring system is proposed for test and evaluation applications. A linear six-state constant velocity (CV) dynamics model is used, consisting of missile position and velocity in a Cartesian coordinate system. Frequency modulated continuous wave (FMCW) radar sensors, carefully located to provide spherical coverage around the target, provide updates of missile kinematic information relative to a drone aircraft. Data from the radar sensors are linearized about a nominal measurement and fused with missile model predictions using an extended Kalman filter (EKF) algorithm. The performance of the system is evaluated through high-fidelity, six-degree of freedom (6DOF) simulations yielding sub-meter end-game accuracy in a variety of scenarios.