Aiming　at　the　incentive　mechanism　issues　in　collaborative　spectrum　sensing　and　access　for　privacy　protection　in　Internet　of　Things　(IoT)　crowd　sensing,　a　multi-task　winner　selection　mechanism　based　on　differential　privacy　is　proposed　to　address　the　privacy　leakage　in　Mobile　Crowd　Sensing　(MCS).　This　mechanism　utilizes　the　exponential　mechanism　of　differential　privacy　to　select　winners,　protecting　the　cost　privacy　information　of　Crowd　Sense　Users　(CSUs).　Furthermore,　to　incentivize　CSUs,　a　two-layer　Stackelberg　game　algorithm　is　proposed.　This　algorithm　can　coordinate　the　compensation　paid　by　Secondary　Users　(SUs)　for　spectrum　access　and　the　data　quality　reported　by　CSUs,　thereby　effectively　enhancing　the　enthusiasm　of　CSUs　to　participate　in　sensing　and　the　performance　of　collaborative　spectrum　detection.　Simulation　results　using　MATLAB　software　show　that　compared　to　traditional　methods,　the　proposed　algorithm　increases　the　average　user　reward　by　19.5%.　©　2025　IEEE.