Resumen de TOPSIS missile target selection method supported by the posterior probability of target recognition

Zhang Jianqiang, Li Jialin, Li Yang

• Addressing the problem of being easily interfered by false targets such as chaff and corner reflector when selecting missiletargets, a multi-feature multi-model target selection method based on the technique for order preference by similarityto ideal solution (TOPSIS) is proposed if only single-target characteristics such as target position or radar cross section(RCS) size is used in the selection of missile targets. The target selection problem under the comprehensive utilisation ofmultiple features is regarded as a problem in decision-making based on multiple attributes. The BP network based on thetarget RCS frequency domain statistical features, the radial basis network based on the target polarisation feature and theradial size feature based on high-resolution range profile (HRRP) are realised. The comprehensive utilisation of C-supportvector machine (SVM) and other target recognition methods and the comprehensive sorting of target selection, to a certainextent, enable anti-ship missiles to make the correct target selection more accurately. The simulation results display thatthe effective selection of ship targets can be achieved in the case of passive interference by surface ships