Efficient One-hop Data Collection Schemes In Wireless Powered Networks

Conventionally,the sensor nodes are usually powered by batteries.However,it is not only laborious but also expensive to replace the nodes' batteries.Also,batteries are inappropriate in some applications due to environmental concerns and the risk of battery leakage.Energy harvesting communication networks can work continuously by harvesting environmental energy,such as solar,wind,thermoelectric effect,etc.However,these traditional techniques can't guarantee the quality of service for the communication networks as the harvested power of these techniques can be highly variable over time.Compared with these traditional methods,the main advantage of harvesting radio frequency(RF)energy is that RF signals are stable,and can carry information and energy simultaneously.Thus,the RF signal is expected to power network nodes in the emerging Internet of Things era.At present,wireless RF power transfer can effectively transmit tens of microwatts of RF power to the destination node from 10 meters away.Wireless powered communication networks(WPCNs)have been a hot research topic in Internet of Things.For WPCNs,in order to obtain a better network performance,it is necessary to simplify the communication and networking protocol of WPCNs as much as possible,so as to reduce the calculation amount of nodes,the energy consumption on calculation and communication.Therefore,the WPCNs using one-hop star topology(hereinafter refer to as star WPCNs)have attracted extensive research interest,they are very important WPCNs.In the star WPCNs,a sink locates in the small-scale network,and coordinates the wireless energy and data transmission with a set of one-hop nodes powered only by the harvested RF energy.Although the design of network protocol of star WPCNs is relatively simple and it is no need to design routing protocol,it still needs to be meticulously designed in energy supply,sink layout,one-hop data collection technology and other aspects to obtain a better network performance.This paper considers the star WPCNs,designs and optimizes the one-hop data collection schemes.The main contents and innovations are:(1)Minimize the network-throughput constrained sink's energy provision(EP).Specifically,this paper investigates the performance of two important MACprotocols: time-division multiple access(TDMA)and non-orthogonalmultiple access(NOMA).For both the TDMA-based WPCN andNOMA-based WPCN,this paper first formulates the EP minimizationproblems as the non-linear optimization problems,then transforms them intoconvex problems,and finally proposes an efficient algorithm which jointlyuses the golden-section search and bisection search methods to determine theoptimal sink's energy transfer and each node's information transmission aswell as the optimal sink's transmit power.Furthermore,for the scenarioswhere the circuit power is negligible,this paper first proves that the optimalsink's transmit power is the maximum allowable value,then provestheoretically that NOMA and TDMA achieve the same EP,and finallypresents a more efficient algorithm for the EP minimization problem.Simulation results demonstrate that TDMA outperforms NOMA when thecircuit power is non-negligible because the circuit energy consumption ofNOMA accounts for a large percentage of the total energy consumption.(2)Investigate the unmanned aerial vehicle enabled WPCN(UAV-WPCN),andminimize the transmission completion time(TCT)of collecting a givennumber of bits per node.This paper transforms this problem into a relativelytractable one by using an area discretization technique so that the nodes'energy harvesting power can be approximated to be the same value under agiven error tolerance ? wherever the UAV is located inside one subregion.Theoptimal solution of transformed problem has an approximation ratio of 1+? tothe theoretically minimum TCT.To solve the transformed problem,this paperformulates it as a covex problem and decomposes it into the master problemand the slave linear programming problem.The master problem is solved by asubgradient based algorithm.Furthermore,for the scenario where each nodehas the same amount of data to transmit,this paper develops an algorithmwith lower complexity.Simulation results demonstrate that the UAV-WPCNgreatly outperforms the conventional WPCN with the fixed sink.(3)Maximize the common throughput of nodes in NOMA-based WPCN.Energytransfer duration,data transmit powers of nodes and the decoding order ofnodes' signals are jointly optimized.This paper first formulates the commonthroughput maximization(CTM)as a non-covex optimization problem,thenseparates the CTM problem into a number of subproblems to solve,andfinally recovers the optimal solution of the original problem from the optimalsolutions of subproblems.Our CTM scheme significantly outperforms theavailable scheme using up all harvested energy especially when the transmitpower of sink is large.Furthermore,this paper proposes an efficient algorithmto maximize the sum throughput while guaranteeing that the maximumcommon throughput is achieved.This scheme further greatly increases thesum throughput especially when the number of nodes is large.