In　this　paper,　we　establish　a　framework　for　low　probability　of　detection　(LPD)　communication　from　a　sequential　change-point　detection　(SCPD)　perspective,　where　a　transmitter,　Alice,　wants　to　hide　her　transmission　to　a　receiver,　Bob,　from　an　adversary,　Willie.　The　new　framework　facilitates　modeling　LPD　communication　and　further　evaluating　its　performance　under　the　condition　that　Willie　has　no　prior　knowledge　about　when　the　transmission　from　Alice　might　start　and　that　Willie　wants　to　determine　the　existence　of　the　communication　as　quickly　as　possible　in　a　real-time　manner.　We　consider　three　different　sequential　tests,　i.e.,　the　Shewhart,　the　cumulative　sum　(CUSUM),　and　the　Shiryaev-Roberts　(SR)　tests,　to　model　Willie＇s　detection　process.　Communication　is　said　to　be　covert　if　it　ceases　before　being　detected　by　Willie　with　high　probability.　Covert　probability　defined　as　the　probability　that　Willie　is　not　alerted　during　Alice＇s　transmission　is　investigated.　We　formulate　an　optimization　problem　aiming　at　finding　the　transmit　power　and　transmission　duration　so　as　to　maximize　the　total　amount　of　information　that　can　be　transmitted　subject　to　a　high　covert　probability.　Under　the　Shewhart　test,　closed-form　approximations　of　the　optimal　solutions　are　derived,　which　will　approximate　the　solutions　obtained　from　exhaustive　search.　As　for　the　CUSUM　and　SR　tests,　we　provide　effective　algorithms　to　search　for　the　optimal　solutions.　Numeric　results　are　presented　to　show　the　performance　of　LPD　communication.