Identification Algorithm Based On Iff And Radar Research

In the modern high-tech war, the identification friend or foe has become increasingcomplexity. Compared with the traditional method of identification, which depends onthe Identification Friend or Foe (IFF) system’s query-response signal, the IFF and radarsystem is the synthesize of the Non-Cooperative and Cooperative target identification.This system can overcome the single sensor’s defect and get the more reliableidentification result than the IFF.The core technologies in the IFF and radar identification system are how to use theradar and IFF’s information to deduce the target’s identity, and how to fuse theidentification results. On the basis of the specific research projects, this dissertationfocuses on the above key technologies and studies the identification algorithms basedon the kenimatic information and IFF’s query-response information. It mainlycomprises:1. Analyzing the uncertainty of the IFF system, get that this system contains therandom uncertainty and fuzzy uncertainty. Based on this, the idea of using the fuzzymathematics and using multi-cycle, various features to deduce the target’s identity canbe obtained.2. Aiming at the problem that how to make the full use of the existingobservational samples and expertise, the algorithm of generating fuzzy rules based onthe data samples has been studied and then the combination of the data samples and theexpertise can be realized.3. Aiming at the problem of using the different kinds of feature information andmultiple sensors to deduce the target’s identity, the hybrid algorithm based on multipleidentification algorithms and D-S evidence theory is proposed. Based on these multipleidentification algorithms, the kinematic information, range profile, IFF’s query-responseinformation can be used to deduce the target’s identity. Based on the D-S evidencetheory, the multiple identification results can be fused.4. Aiming at the problem that how to update the identification result and how tomake the observation’s degree of confidence input to the identification algorithm, the algorithm which combines the discrete fuzzy dynamitic Bayesian Network, multi-layerfuzzy comprehensive evaluation and D-S theory is proposed. Based on the discretefuzzy dynamitic Bayesian Network, the identification result can be updated with theobservations’ augment. Based on the multi-layer fuzzy comprehensive evaluation, themultiple observations and radars’ degrees of confidence can be taken into account.Based on the D-S evidence theory, the multiple identification results can be fused. Theradar and IFF system can obtain the more reliable result based on this algorithm thanusing the single sensor.The effectivenesses of the methods are validated by simulations. On the conditionthat there are only observational samples and expertise, using the proposed methods canaccomplish the identification friend or foe based on the multi-cycle, different kinds offeature information, multiple radars and IFF.