Research On Simulation Platform Of Data Fusion And Related Algorithm In Multi-radar Networking

Stealth targets, such as F22, has characteristics of invisibility and strongmaneuverability, which is a big challenge to build the new generation of homeland airdefense early warning net. Multi-radar networking can detect airspace from differentangles, and get multi-dimensional battlefield information. Comprehensive utilizationof information fusion technology can improve the ability to detect this kind of stealthtargets.To verify the performance of information fusion algorithm and manage networkresources effectively, this thesis designs a multi-radar networking simulation platformbased on the decoupling structure of information fusion and resource managementwith feedback, realizes main functional modules, and performs research onanti-stealth radar networking optimization deployment issues. The main works are asfollows:Firstly，the research background and significance are introduced, and currentprogress of radar networking and distributed simulation technology at home andabroad are reviewed.Secondly, the framework of simulation system based on high level architecture isdesigned through analyzing the requirements of radar networking data fusion system.The main federal members of the federation include target, radar, fusion center，sensormanagement, etc. The function and input/output interface of each federal member isdescribed, and the multi-radar networking data fusion simulation platform isestablished based on Mak RTI atfer realizing each module’s function. The relatedalgorithms are transplanted and tested.Thirdly, the optimization deployment of miti-stealth multi-radar networking isstudied in order to realize the static management of radar networking. The simplifiedmodel of single radar to detect stealth target is established, and a two-leveloptimization index is built for anti-stealth background. An optimization algorithmcombined PSO with Powell is proposed to resolve the netted radar deployment optimization problem. Simulation results verify the validity of the proposed algorithmwhen compared with PSO algorithm.Finally, the current work is summarized.