Large-scale Simulation System Of Hla-based Data Compression And Transmission Technology

The main research work of this paper is data compression and transport technique (applied) in HLA-based massive simulation systems. HLA has been the standard structure of distributed simulation system. However, as the primary undertaker of data interaction and network real-time quality in HLA system, RTI cannot accomplish the transport of certain data types very well, or fails to meet the requirement of real-time transport. Based on the analysis of the inadaptability of high volume data transport in nowadays HLA/RTI systems, the paper presented an approach for this problem. In our approach, token data, such as data size, data type, file name, storage path, updated time etc., were sent via RTI, and used to accomplish data publishing and subscribing, while the content of data went through the network layer. Further, we designed a relevant system framework based on this approach.For the purpose of improving transport efficiency, we need to introduce data compression technique into the HLA system. In this paper, we designed a function module for data compression and transport. Aimed at some interaction data types possibly used in the HLA system, this module chose and implemented some typical compression algorithm, and encapsulated functions which use TCP/IP and UDP/IP responsive/non-responsive mode to accomplish data transport. By embedding this module into HLA federate, we solved the problem of combining these two techniques with HLA system.In the end of the paper, a simple experiment of HLA simulation model testified the feasibility and validity of our engineering method. From the integrated result, the average transport delay can be controlled less than 100ms, and can satisfy the system's real-time requirement.