Researches On Data Distribution Service Transmission Optimization In Limited Bandwidth Networks Environments

Maritime combat environoment of modern information war is getting more complex, thus raises higher requirements on data communication. As a primary communication network connecting combat units, wireless network takes on narrow bandwidth,low transfer rate, and long transceiver conversion time.Current military applications mainly adopt IP-based data distribution service for data communication, but this kind of service is of low commuication efficiency and bandwidth utilization, and hard to meet the requirements of efficient and real-time data transmission on military applications. To solve these problems, this paper puts forward information cutting technique for interacting entities and compressed transmission technique, optimizing the data distribution service in terms of the publish-subcribe entity discovery process and transmission process individually.Based on detailed analysis and research on real-time publish-subcribe protocols, the information cutting technique for interacting entities optimizes information interaction processes among the domain participants and endpoint. According to the characteristics of the target applications, this technique stores in XML files the domain participants interaction and endpoint interactive informations which are known in advance or unchanged, optimizes entity discovery process by static configuration files and dynamic data transfer, and speeds up the entity discovery process by the way of eliminating unnecessary data transmission in discovery process, on the premise of guaranteeing success of basic publication and subscription process.The compressed transmission technique compresses data before sending upon UDPv4 protocols, making use of compression algorithms and transfer plug-ins, which further reduces the amount of data, and increases utilization rate of network bandwidth.The domain participants finding plug-in for cutting interacting information, endpoint finding plug-in, and compression transmission plug-in were implemented. Testing scheme was set up and the DDS transmission performance after using these optimized plug-ins was carried out. With the application of domain participants finding plug-in, the amount of domain participants interaction information is reduced about 44.4% to 53.4%., the discovery time of entities is reduced about 51.1%. After applying endpoint finding plug-in, the endpoint interaction information is reduced about 29.9% and the discovery time about 69.5%. The compression efficiency of applying the Zlib compression transmission plug-in is from 27.1% to 60.4% and the compression efficiency of applying the Bzip2 compression transmission plug-in is from 16.6% to 66.7%, which improves the utilization rate of network bandwith.The optimization techniques realized in this paper has been applied in actual projects.