Community Service Selection Algorithm for Network Simulation Task

Net-centric simulation (NCS) is a new simulation mode, which provides a brand new method to support distributed simulation in WAN. To meet different users' requirements, building a new simulation task community (STC) dynamically by integrating distributed various services on network is the key problem of NCS. A simulation task community service selection algorithm (STCSSA) is proposed. It aims at the building of STC and adopts an improved particle swarm optimization (PSO) which overcomes the shortcomings of traditional PSO. The essence of the algorithm is that the problem of STC construction is transformed into a multi-objective service selection optimization with QoS constraints. Firstly, QoS model of STC service and the evaluation of STC composition process is described. Then, STCSSA is introduced in terms of the function steps, including a dynamic inertia weight strategy and an alternative method of mutation. Finally, several experiments are conducted to test STCSSA. On the one hand, STCSSA is tested with typical functions to prove the proposed algorithm can not only improve the convergence speed of service options, but also avoid the algorithm going into a local optimum. On the other hand, practical application is presented in detail. STCSSA is utilized to help a military STC search simulation services. Compared with some other PSO algorithms, STCSSA can effectively support large-scaled STC construction in NCS environment.