Simultaneous　wireless　information　and　power　transfer　is　an　innovative　way　to　provide　electrical　energy　for　user　equipments.　However,　most　previous　works　mainly　focus　on　energy　harvesting　over　a　relatively　narrow　frequency　range.　Due　to　small　energy　harvested　by　the　users,　the　practical　applications　are　usually　limited　to　low　power　devices.　In　this　paper,　an　energy-efficient　uplink　resource　allocation　problem　is　investigated　in　a　cellular　network　with　ambient　radio　frequency　(RF)　energy　harvesting.　In　order　to　obtain　sufficient　energy,　a　broadband　rectenna　is　adopted　to　harvest　ambient　RF　energy　over　six　frequency　bands　at　the　same　time.　From　the　viewpoint　of　service　arrival　in　the　ambient　transmitter,　a　new　energy　arrival　model　is　presented.　The　problem　of　sub-carrier　and　power　allocation　is　formulated　as　a　mixed-integer　nonlinear　programming　problem.　The　objective　is　to　maximize　the　energy　efficiency　while　satisfying　the　energy　causality　and　the　total　data　rate　requirement.　In　order　to　reduce　the　computational　complexity,　a　suboptimal　solution　is　derived　by　employing　a　quantum-behaved　particle　swarm　optimization　(QPSO)　algorithm.　Simulation　results　show　that　the　QPSO　method　has　higher　energy　efficiency　than　a　conventional　particle　swarm　optimization　approach.