On The Performance Optimization Of RF-energy Powered Communication Networks

As predicted by the GSMA,the number of connected electronic devices will be soaring up to 25 billion by 2025,which gives birth to a novel communication paradigm —the Internet of Things(IoT).With such a tremendous amount of IoT devices,the limited life time of batterypowered networks becomes a critical concern due to the immense cost of battery replacement and environmental issues.Recently,radio-frequency(RF)energy has been approved as a potential power for IoT devices.Via introducing radio-frequency(RF)energy harvesting(EH)technology into IoT,the newly-emerged RF energy powered communication networks has drawn a lot of attention.With the RF-based EH technology,the lifetime of networks can be greatly extended and even could be unlimited.However,some new challenges(e.g.,the slow harvesting rate of RF energy,the significant attenuation during RF energy transfer,the long-term system operation and the highly channel-state-related system performance)are posed and it is of great importance to find new solutions for the optimization problems in RF energy powered communication paradigm.In this paper,we devote our efforts to the optimal design of RF energy powered communication systems.Specifically,we aim at the node position optimization in ambient RF energy powered communication systems,the energy efficiency maximization of a heterogeneous wireless powered communication networks(WPCN),the long-term throughput maximization of a single-user WPCN,and the energy saving dispatch strategy of the mobile sink in a clustered-WPCN.Our main contributions are summarized as follow.1)Considering an ambient RF energy powered communication system in which sensor nodes are powered by the ambient RF signal emitted from an existing wireless communication system,the outage performance of a point-to-point sensor communication link is first analyzed.Due to the scarce spectrum resource,the ambient RF signal is not only considered as an energy source but also the in-band interference for the sensor communication link.Besides,to counter the slow harvesting rate of ambient RF energy and ensure adequate system energy supply,the sensor node is designed to harvest RF energy in multiple time blocks.Via analyzing the signal to interference and noise ratio(SINR)of the system,the sink node position is optimized to reduce the link outage.After that,we extend the work into a DF relay scenario and provide a sub-optimal solution for selecting relay position.Finally,all the theoretical results are verified by numerical simulations.2)A heterogeneous WPCN which consists of a hybrid access point(H-AP),several user equipments(UE)and a device-to-device(D2D)communication pair(DT and DR)is investigated.According to the service type,the network is considered as a combination of two subsystems(UE-WPCN and D2D-WPCN)and all the UEs and the D2 D pair are powered by the RF energy transferred from the H-AP.Aiming at maximizing the energy efficiency of the network subject to given QoS requirements of these two sub-systems,the transmit power of the H-AP,the wireless energy transfer duration,the transmit power and transmission time of the UEs and the DT,as well as the spectrum allocation between the two sub-systems are jointly optimized based on nonlinear fractional programming and convex optimization.Simulation results show that the proposed solution outperforms the existing throughput-oriented solution in terms of energy efficiency.The impact of the transmit power of the H-AP,the energy transfer efficiency and the circuit power on the system performance are also investigated.3)For the circumstance where the RF energy transfer technology supports sustainable data communications,a single-user WPCN operating in an infinite horizon is considered.Aiming at maximizing the long-term system throughput with a limited energy budget,we devote our effort to the online policies design of the system.Via introducing the theory of constrained Markov decision process(CMDP),an optimal online transmission policy is devised for scheduling the H-AP and the UE.After that,by mimicking the properties of the optimal offline policy,a suboptimal online policy is proposed to cut down the complexity.The simulation results show that the sub-optimal online policy significantly reduces the computation time but reaches comparable performance to the optimal online policy.4)To improve the RF energy transfer efficiency in large-scale applications,we design a clustered structure for WPCNs via introducing dual-functional cluster heads(CH)and a versatile mobile sink(MS).The CH can relay the MS's RF energy to the inner-cluster nodes and deliver nodes' data to the MS.Based on the existing TSP-based solution,an energy saving dispatch strategy is devised for the MS via utilizing the dynamic programming(DP)theory,so as to cut down the network energy consumption.Besides,a position-adjustment-oriented heuristic dispatch strategy is also proposed to reduce the complexity.The simulation results show that,compared with the existing TSP-based solution,the network energy is considerably reduced by adopting the two proposed dispatch strategy.