In OFDM-based System such as long term evolution(LTE),the scheduling scheme plays an essential role in not only improving the capacity of system,but also guarantee the fairness among the user equipments(UEs).However,most existing work about scheduling only considers the current throughput in physical layer.Thus in this paper,a cross-layer scheduling with fairness based on restless bandit(CSFRB)scheme with the ’indexability’ property is proposed for the multi-user orthogonal frequency-division multiplexing(OFDM)system to minimize the distortion in the application layer,to maximize the throughput and to minimize the energy consumption in the physical layer.The scheduling problem is firstly established as a restless bandit problem,which is solved by the primal-dual index heuristic algorithm based on the first order relaxation with low complexity,to yield the CSFRB scheme.Additionally,this scheme is divided into offline computation and online selection,where main work will be finished in former one so as to decrease the complexity further.Finally,extensive simulation results illustrate the significant performance improvement of the proposed CSFRB scheme compared to the existing one in different scenarios. In OFDM-based System such as long term evolution (LTE), the scheduling scheme plays an essential role in not only improving the capacity of system, but also guarantee the fairness among the user equipments (UEs) .However, most existing work about scheduling only considers the current throughput in physical layer. Tan in this paper, a cross-layer scheduling with fairness based on restless bandit (CSFRB) scheme with the ’indexability’ property is proposed for the multi-user orthogonal frequency-division multiplexing (OFDM) system to minimize the distortion in the application layer, to maximize the throughput and to minimize the energy consumption in the physical layer. The scheduling problem is established as a restless bandit problem, which is solved by the primal-dual index heuristic algorithm based on the first order relaxation with low complexity, to yield the CSFRB scheme. Additionally, this scheme is divided into online computation and online selection, where main work will be finished in f ormer one so as to decrease the complexity further. Finally, extensive simulation results illustrate the significant performance improvement of the proposed CSFRB scheme compared to the existing one in different scenarios.