Maximum Throughput of a Cooperative Energy Harvesting Cognitive Radio\n User

In this paper, we investigate the maximum throughput of a saturated\nrechargeable secondary user (SU) sharing the spectrum with a primary user (PU).\nThe SU harvests energy packets (tokens) from the environment with a certain\nharvesting rate. All transmitters are assumed to have data buffers to store the\nincoming data packets. In addition to its own traffic buffer, the SU has a\nbuffer for storing the admitted primary packets for relaying; and a buffer for\nstoring the energy tokens harvested from the environment. We propose a new\ncooperative cognitive relaying protocol that allows the SU to relay a fraction\nof the undelivered primary packets. We consider an interference channel model\n(or a multipacket reception (MPR) channel model), where concurrent\ntransmissions can survive from interference with certain probability\ncharacterized by the complement of channel outages. The proposed protocol\nexploits the primary queue burstiness and receivers' MPR capability. In\naddition, it efficiently expends the secondary energy tokens under the\nobjective of secondary throughput maximization. Our numerical results show the\nbenefits of cooperation, receivers' MPR capability, and secondary energy queue\narrival rate on the system performance from a network layer standpoint.\n