Research On Protocol Derivation Based On Surveillance State

With rapid development of wireless data link system, Unmanned Aerial Vehicle (UAV) has been widely used in many areas. Meanwhile, how to enhance the reliable measuring and control of UAV is becoming an issue that is urgently solved. Protocol packets are the payload of wireless data link. Completing the inference work of non-public protocol packets will make an important effect on communication protocol design, electronic countermeasure and strengthening of security of UAV. Aiming at the outside state parameters that is observed by radar, basing on the network trace and combining the real time change of UAV outside state parameters, this paper has designed an inference system adapting to data link application layer binary system protocol, borrowing ideas from execution trace. Here are following research work done:(1) Measurement and control protocols are defined and communication message are produced. According to fashionable data link system and flying control features of UAV, this article systematicly define communication protocol model that can be used on measurement and control field, referring to ways like ABNF. Meanwhile, protocol messages library based on protocol model is produced by means of simulating the communication process during the measuring and control.(2) Frame features and domain structure relations are obtained. Pattern string matching is used to seek packets’frequent pattern string and frequent position; spare frame feature set is composed; association rules algorithm is used to dig and select association rules in spare frame feature set; the protocol structure is analyzed according to association relations of frame features; the domain structure relation sets and packets information of being classified are obtained.(3) The state parameter data domain is analyzed. The external state parameters are used to analyze state data domains of packets sequence and search key structures in state data domains according to the characteristics of binary protocol. State data domains’structure relations are used to analyze and confirm attribute information of each unit in domain structure relations through searching results and comparing results.(4) Context constraints based on behavior features are identified. Common behavior features in most protocol conversation and domain are used to analyze all packets in protocol message library and find context constraints possessing behavior features and domain being effected. The behavior features met is used to infer attribute information that domain possesses.Packets and state parameters are taken as input and a completed protocol inference system is designed. Protocols message library and flying state parameters are used to test inference system.Finally effective inference of all packets patterns in protocol message library is realized by way of constantly modifying in each step.