Traffic　monitoring　is　a　crucial　means　to　solve　traffic　problems　by　collecting　vehicular　datain　an　urban　road　network.　However,　the　leakage　of　sensitive　data　increases　the　risk　of　attackers　spying　on　drivers＇　privacy.　To　achieve　real-time　traffic　monitoring　without　leaking　data　privacy,　we　propose　a　privacy-preserving　compressive　sensing　(PPCS)　mechanism　to　timely　achieve　the　entire　road　network　traffic　conditions　and　design　a　series　of　feasible　and　securecomputation　protocols　to　accelerate　the　speed　of　data　processing.　Specifically,　we　utilize　the　cloud　to　extract　the　spatial-temporal　correlation　between　different　road　segments　and　compute　a　small　number　of　road　segments＇　traffic　congestion　rates.　Notably,　we　adopt　compressivesensing　(CS)　technology　to　estimate　other　unknown　road　segments＇　traffic　conditions　based　onseveral　known　values.　Besides,　a　comprehensive　security　analysis　indicates　that　the　PPCS　proposed　can　protect　vehicular　data　from　being　disclosed.　Finally,　we　validate　the　effectiveness　and　efficiency　of　the　mechanism　through　simulated　experiments　of　computational　costs　and　communication　overhead.