Random finite set based data assimilation algorithm for dynamic data driven simulation

Computer simulation has long been used for studying and predicting behaviors of complex systems. With the recent advances in sensor and network technologies, the availability and fidelity of real time measurements have been greatly increased. This makes the new simulation paradigm of dynamic data driven simulation more and more popular. It can assimilate the real time measurements for much better analysis and prediction of complex systems. Data assimilation techniques are the foundation of the dynamic data driven simulation, but the traditional particle filter based data assimilation algorithms can't meet the actual application requirements. In this paper, we study how to utilize the real time measurements for the dynamic data driven simulation. A new random finite set based data assimilation algorithm is proposed to overcome the limitations of the standard data assimilation algorithms. The random finite set based measurement model and simulation model that are used in the data assimilation process are introduced. The detailed implementation of the random finite set based data assimilation algorithm is presented. The study case with anti-piracy is used to practically illustrate the proposed data assimilation algorithm. The effectiveness and accuracy of the algorithm are checked by experiments.