Nowadays,　critical　infrastructures　are　essential　for　the　normal　operation　of　economics　and　social　life.　Because　of　the　increasing　number　of　terrorist　attacks　and　other　accidents,　research　on　strengthening　infrastructure　protection　has　become　essential.　Meanwhile,　the　operation　of　many　types　of　infrastructures　mainly　relies　on　the　configuration　of　networks.　The　interdependent　infrastructure　network　(IIN),　which　is　a　network　comprising　multiple　infrastructure　networks　with　different　functions　under　supporting　relationships,　is　highly　influenced　by　the　interdependency　structure　of　these　networks　and　the　protection　strategy.　To　better　understand　the　relation　between　the　interdependency　structure　and　protection　in　two　phases　on　infrastructure　resilience　and　analyze　their　influence　on　IIN　performance,　this　study　investigates　the　IIN　in　a　community　and　proposes　a　joint　optimization　method　for　the　two　phases　based　on　an　incomplete-information　game　under　the　situation　of　intentional　attacks.　Employing　the　maximal-flow　index　(MFI)　as　the　performance　metric,　a　multi-objective　planning　approach　is　proposed　for　the　joint　optimization　of　interdependency　structure　and　protection.　Furthermore,　an　improved　probabilistic　solution　discovery　algorithm　(PSDA)　is　developed　to　solve　the　planning　model,　and　the　joint　optimization　is　analyzed　using　extensive　experimentation,　including　three　real-world　infrastructure　examples.