Security　threats　to　smart　and　autonomous　vehicles　cause　potential　consequences　such　as　traffic　accidents,　economically　damaging　traffic　jams,　hijacking,　motivating　to　wrong　routes,　and　financial　losses　for　businesses　and　governments.　Smart　and　autonomous　vehicles　are　connected　wirelessly,　which　are　more　attracted　for　attackers　due　to　the　open　nature　of　wireless　communication.　One　of　the　problems　is　the　rogue　attack,　in　which　the　attacker　pretends　to　be　a　legitimate　user　or　access　point　by　utilizing　fake　identity.　To　figure　out　the　problem　of　a　rogue　attack,　we　propose　a　reinforcement　learning　algorithm　to　identify　rogue　nodes　by　exploiting　the　channel　state　information　of　the　communication　link.　We　consider　the　communication　link　between　vehicle-to-vehicle,　and　vehicle-to-infrastructure.　We　evaluate　the　performance　of　our　proposed　technique　by　measuring　the　rogue　attack　probability,　false　alarm　rate　(FAR),　mis-detection　rate　(MDR),　and　utility　function　of　a　receiver　based　on　the　test　threshold　values　of　reinforcement　learning　algorithm.　The　results　show　that　the　FAR　and　MDR　are　decreased　significantly　by　selecting　an　appropriate　threshold　value　in　order　to　improve　the　receiver＇s　utility.