Study Of Identification Friend Or Foe Based On Information Fusion Of Radar And IFF

As traditional cooperative identification friend or foe method, like identificationfriend or foe system (IFF), obtains limited target identities and has high requirement ofsystem reliability, target identity can’t be discriminated accurately enough. Modernwarfare is information-based high-tech warfare, and its environment is full of all kindsof interference. For this reason, obtaining reliable target identity is the premise in orderto win the war. However, traditional cooperative identification friend or foe methodcan’t meet the demand, so it’s necessary to build an integrated identification networkcombining with non-cooperative identification method, to enlarge the coveredtime/space/frequency domain of the system and discriminate more reliable targetidentity. High efficient and reliable information fusion technology is the key point toguarantee the accuracy of this system.Supported by the issue of integrated identification friend or foe based on radars andIFF, this thesis mainly study uncertainty processing, multi-period fusion algorithm,modified D-S algorithm which can resolve conflict among computed identity results ofrules and the frame of integrated identification friend or foe. The main content of thisthesis include:1. Analyze uncertainty existing in radar and IFF of the system, then selectappropriate identity inference algorithms and fusion methods according to the previousanalysis, to reduce the influence of fuzzy uncertainty and random uncertainty on thesystem output.2. Study multi-period fusion methods. Analyze the over-estimating problem usingD-S evidence theory to fuse multiple periods firstly. Then propose a period fusionmethod based on fuzzy comprehensive evaluation, and prove it can remove part ofwrong ID deductions of single periods and accelerate fusion by simulation.3. Study evidence reform method based on evidence discount that can partlyreduce conflict, which remains D-S combination rules in favor of engineeringfulfillment. Come up with an estimation method of fusion result credibility based on revised evidence sets, and prove the modified D-S has the ability of conflict resolutionafterwards, finally confirm the proposed credibility estimation method is reasonable.4. Build a frame of integrated identification friend or foe based on radars and IFFwith existing identification friend or foe methods, proposed multi-period fusionalgorithm and modified D-S solving conflict. Besides, the realization of each module isintroduced in detail, and proper update patterns of these modules are suggested. Finally,experiments based on constructed track data verify the advantages and expandability ofthis frame.