The Research And Implementation Of An Aircraft Simulation Training System With QoS

As the development of networks, communications and the simulation technology, distributed simulation technology is used in military field more and more, especially in the field of large-scale distributed interactive simulation system. The efficiency of network communication as the supporting foundation constraints the efficiency of the simulation system directly, so the real-time distributed simulation system needs a very stringent requirement to the communication.HLA (High Level Architecture) has become the main structure of the real-time distributed simulation system. And the management of communication is implemented by RTI. But RTI is mainly used to implement the management of communication between a number of independent components or simulation systems. But there are not enough concrete solutions to complex client simulation system, but it is important to distinguish the communication requisitions for the simulation system at Client side. In addition, how to simplify the design and development of decision-making logic module in Server side is a major problem of a large-scale distributed simulation system, because the development of the decision-making logic modules is related to many fields of knowledge, but our professional engineers' computer capability are different, and logistic error will be appeared in program easily without a unified model supporting.Against the problem above, on the basis of the design and implementation process of the distributed simulation system, the thesis focuses on two key aspects of study:One is to discuss the mechanism of communication of distributed simulation, and the asymmetric communication mechanism between the components and content-distinction based IP request queue scheduling are proposed to ensure the QoS of the communication requirement from the complex simulation system in Client, thus it improves the accuracy of the simulation. Another aspect is the design of a logic decision tree model in Server to simplify the complex logic decision-making. According to the logic decision tree model built, it greatly reduces the difficulty and the risk of errors in preparing the decision logic.