Research of service selection algorithm for net-centric simulation task community

Abstract

Being a new-generation C4ISR system simulation method, the construction approach of net-centric simulation (NCS) is developing toward net-centric from the traditional approach of platform-centric. NCS is mainly completed by the construction of the simulation task community (STC), the key to which being the dynamic integration of the various services spread in the network in order to form a new STC that meets the requirements of different users. In this study, a simulation task community service selection algorithm (STCSSA) is proposed. The main idea of this algorithm is to transform the construction of STC to the searching of optimal multi-objectives services with QoS global constraints. This paper first introduces the QoS model of STC and evaluates the service composition process, then presents the detailed operating process of STCSSA and design of the dynamic inertia weight strategy of the algorithm, and also proposes an optional variation method. Comparative tests were performed on STCSSA with other particle swarm optimization algorithms. It was validated from the perspective of performance that the proposed algorithm has advantages in improving the rate of convergence and avoiding local optimum, and from the perspective of practical application STCSSA also demonstrated feasibility in the construction of large-scale NCS task community.