A fully distributed data collection method for HLA based distributed simulations

Abstract

The data collection in simulations is important especially for the use in After Action Review (AAR). In distributed simulations there is still not a standardized solution for data collection. The main reason is the diversity of the simulations and the new developments in this area. It was less complicated to collect simulation data with DIS standard; but the relatively new standard, HLA, not only has come with the optimizations, reusability and more interoperability but also with the complication and overhead of data collection. With the use of HLA, the simulation's state data has got along together more in simulators/federates own domain rather than being mitted to the simulation network. So, a solution for collecting this important private data should also be introduced. We first examine the need of data collection, especially in AAR; then overview the possible solutions, with their pros and cons and argue how a robust data collection can be achieved. After that, the proposed and implemented architecture is described. The paper ends with the benchmarking results. © 2009 Simulation councils, inc. © 2012 Elsevier B.V., All rights reserved.