In　order　to　solve　the　problem　of　fault　diagnosis　method　for　analog　IC　diagnosis　based　on　neural　network,　the　method　based　on　particle　swarm　optimization　(PSO)　and　least　squares　support　vector　machine　(LSSVM)　is　proposed.　Wavelet　package　is　used　as　a　tool　for　extracting　feature.　Then,　after　training　the　LSSVM　by　PSO,　the　model　of　the　circuit　with　fault　diagnosis　system　is　built.　Simulation　results　show　that　the　method　is　more　effective.