Reseach On Self-organizing Multi-source Coordination Detection Technongy

Multi-source coordination detetion system, which is base on self-organizing network and data integration methods, is progressively used in so many fields of military and civil affairs, such as forest fire early warning,urban monitoring radar scheduling and missile tracking. A large number of equal mobile network nodes that have the capacity of communication, computing, storage spontaneously form the self-organizing sensor network according to the distributed network protocol. Each node can collecte, calculate, decide information independently, and also can communicate, fuse information, collaborate with each other. Therefore, multi-source detecting system has stronger robustness, self-healing ability, information processing ability, and a wider coverage of space, more flexible and efficient search, location and tracking ability.In this paper, the study of multi-source coordination detetion system will include the following content: coverage is the the precondition of the subsequent task, the effective coverage area must be meet standards; sensor management is the technological mean to carry out tasks and scientific and effective management methods can not only improve the detection performance but also extend the service life at the same time; in a particular case, dynamic coordination tracking is needed to improve the whole system tracking performance.The main contents are:1、In order to improve the deployment algorithm which based on virtual force, a clustering dynamic deployment algorithm is proposed.The algorithm has improved the problem that sensors are concentrated in the center and hard to disperse at beginning.2、Data fusion subsystem and sensor management subsystem are connected with each other, a new sensor management model is proposed through combine regulation to reduce the calculation capacity of the original algorithm.3、Considering the coordinated tracking problem, the information entropy is powerless only if increase distance and angle information. A new concept, unit information entropy increment is defined, which can be used to control sensor motion and improve the tracking ability.