A　Sybil　attack　is　caused　by　a　malicious　vehicle　node　stealing　fake　identities　and　continuously　generating　fake　vehicles　on　the　road　to　create　the　illusion　of　congestion,　which　endangers　normal　vehicles　on　the　road.　Because　Sybil　vehicle　nodes　have　trajectories　and　motion　states　similar　to　those　of　normal　vehicles,　they　are　more　difficult　to　detect　in　high-density　traffic　environments.　The　real-time　authentication　of　vehicles　is　impossible　in　the　existing　traffic　environment;　thus,　malicious　vehicle　nodes　with　a　high　degree　of　stealth　can　continuously　attack　and　are　difficult　to　stop.　In　this　paper,　we　propose　a　Sybil　attack　detection　method　based　on　basic　security　message　(BSM)　packets,　which　exploits　the　characteristic　that　BSM　packets　have　a　unique　sending　source　and　uses　the　spatiotemporal　relationship　of　dynamic　vehicle　location　changes　to　detect　and　trace　Sybil　attacks.　A　weighted　integration　strategy　is　proposed　for　increasing　the　detection　precision　without　machine-learning　model　prediction.　Experimental　results　indicated　that　the　proposed　method　can　detect　Sybil　attacks　in　real　time　and　is　not　affected　by　the　attack　density　or　traffic　density.　Moreover,　it　can　detect　Sybil　nodes　and　trace　malicious　nodes　simultaneously,　with　precisions　of　＞98%　and　＞94%,　respectively,　resolving　the　difficulties　of　existing　detection　schemes.