Data Transmission Schemes With Optimization For Battery-Free Wireless Sensor Network

With the further development and application of Internet of things(IoT),data collection and delivery for the physical world are becoming increasingly important.Wireless Sensor Network(WSN)palys an important role of collecting and delivering data in the sensing layer of the IoT.Many applications(such as building structure monitoring,human health monitoring,etc.)require that the nodes can work for a long time after deployment.Battery-powered WSN is not qualified for these applications due to the fact that batteries need to be replaced frequently.This shortcoming can be solved by Battery-Free Wireless Sensor Network(BF-WSN),in which the nodes harvest energy from sunlight,vibration,wind,radio frequency(RF)signals,or other sources,so that the nodes can be provided with sustainable energy,which guarantees a long period of data collection and delivery for the IoT sensing layer.Energy harvesting in the nodes depends on the ambience.Data collection and delivery in the BF-WSN involve many elements,such as energy harvesting rate,charging time,residual energy,energy consumption level,transmission delay,and reliability of data transmission.As a result,how to integrete these elements to design appropriate data delivery schemes is very important and challenging.This paper focuses on this topic,which contains the main contributions and innovations as follows:1.The problem of low latency and reliable IPv6 packet delivery over BF-WSN is studied,which yields the Latency Aware IPv6 Packet Delivery(LAID)scheme to reduce the IPv6 packet latency while maintaining high packet delivery ratio(PDR).LAID considers charging time,data rate,and the Maximum Number of Transmission Trials(MNTT)so that the minimum latency can be achieved by optimizing the pairing of data rate and MNTT.In addition,network coding is applied to improve IPv6 packet delivery reliability.For the LAID scheme,we define its time line,the preparation time of nodes,and the data transmission time.The successful transmission probability of a MAC(Medium Access Control)frame,the average transmission time of an IPv6 packet,the average energy consumption of an IPv6 packet,and the end-to-end latency over the route are derived based on the Bit.Error Ratio(BER)model using the modulation mode of Offset Quadrature Phase Shift Keying(O-QPSK).The preparation time of nodes are derived base on the instantaneous energy harvesting rate and the average receive energy consumption of an IPv6 packet.The optimization problem(OP)aiming at minimizing the delivery delay(including the preparation time and the IPv6 packet transmission time)of each hop over the route is given,and the data rate and MNTT are optimized.Theoretical analysis and simulation results show that LAID can significantly reduce the latency of IPv6 packet.2.The problem of reliable data collection in BF-WSN is studied,which leads to the Scheme of Accumulating Correct Data Blocks(SACDB)for the RFP-WSN(Radio Frequency Powered Wireless Sensor Network).With the SACDB,a node generates a packet composed of multiple data blocks with each having its own check so that the sink can accumulate the correct blocks in the erred(i.e.,bit errors)packets,which reduces the number of retransmissions of a packet and improves the successful transmission probability of packets over the unreliable wireless links.In addiction,the MAC frame format compatible with IEEE 802.15.4 standard is designed,and the operation steps of the sink and the nodes are given.For the SACDB scheme,a mathematical analysis model based on node state transition probability is proposed.The successful transmission probability of a packet,the average energy consumption of a packet,and the average delay of a packet are derived by using the k step state transition probability matrix and the C-K(Chapman-Kolmogorov)equation.Then the effective throughput of a node is derived,and the OP with maximum the effective throughput of a node is formed.The optimal number of blocks in a packet is obtained by solving the OP.Simulation results show that the SACDB outperforms the existing schemes in throughput,packet delivery ratio(PDR)and energy efficiency.3.The problem of energy-efficient data collection in BF-WSN is studied based on the above SACDB scheme,which leads to the Radio Frequency-powered Data Collection Scheme(RF-DCS).The RF-DCS which extends the SACDB takes channel allocation,energy threshold and packet block-level retransmission mechanism into account.The values of the two parameters of the number of blocks in a packet and the enery threshold are optimized jointly,and the network energy efficiency and throughput are derived.In addition,a network utility function that can integrate network energy efficiency and throughput is defined,and the OP that maximizes network utility is formed.By adjusting the utility coefficient and solving the OP,the optimal energy threshold and the optimal number of blocks per packet can be obtained,which are used to maximize the energy efficiency and throughput.Simulation results show that the energy efficiency and throughput of the RF-DCS are higher than that of the existing schemes.The above schemes that deliver data to the destination with reliable,low power,and high throughput can be applied to the BF-WSN,among which the LAID scheme is suitable for delivering IPv6 packets over BF-WSN,and the SACDB and RF-DCS schemes are suitable for the collection of the sensed data in the scenario of radio frequency energy harvesting.The results of this paper have important theoretical significance and practical value.In addition,the BF-WSN nodes are powered not by batteries but by the energy harvested from the ambience,which are conducive to energy saving,emission reduction,and protection of the ecological environment.