Innovation in the field of intelligent autonomous systems of the automotive sector has been ever increasing. Accurate tracking of vehicles is an important aspect in the design of applications such as smart route planning or collision avoidance systems. In practical applications, tracking of vehicle using radar technology suffers from serious problem due to noisy measurements. It introduces major limit on the accuracy of the tracking system. This paper discusses a case study scenario where the robustness of vehicle tracking can be improved using Extended Kalman Filtering. Noisy radar measurement is simulated through model based design (MBD) using MATLAB. Analysis and design of Extended Kalman Filter to mitigate the noise is discussed. An efficient system architecture to implement the algorithm in autonomous smart vehicle tracking system is also identified.