The Research On Data Exchange Mechanism Of Low-power Wireless Transmission System

With the advent of the "Industrial 4.0 Era",the Internet of Things(IOT)technology is becoming more and more mature,gradually linking the "world of things" to the Internet.Nodes in the IOT need to transmit sensing data in real time.In practical use,low-power wireless transmission technology plays a vital role because of the size and energy constraints.The efficiency of data exchange in low power wireless transmission system always affects the transmission performance of the whole system.Therefore,how to further improve the data exchange efficiency of low-power wireless transmission system in the IOT is very important.The main work of this dissertation is to improve data exchange efficiency in wireless transmission system.The main work is as follows:(1)Uplink data collection method for passive nodes based on parallel decoding is proposed.In the scenario of large passive nodes deployment,the nodes are inefficient in single-slot communication,and parallel transmission can improve the efficiency of data collection.In existing methods,nodes are modulated in Amplitude Shift Keying,which has poor anti-jamming performance.Receiver decoding needs to rely on stable channel coefficients.In this dissertation,the modulation mode of the node is optimized.The receiver maps the node information into the constellation domain and decodes the node in parallel according to the nodes' state and the sequence characteristics of the cluster.In this work,compared with Amplitude Shift Keying modulation,Phase Shift Keying modulation of nodes is more obvious in constellation domain clustering.Compared with existing methods,this protocol improves the accuracy of parallel transmission and decoding of passive nodes.(2)Downlink data feedback protocol for passive nodes based on link characteristic aware is designed.In existing data feedback protocols,the receiver can only perform data feedback with a single node physical connection.If multiple nodes in a unit area perform the same feedback to distribute data,the receiver still needs to communicate with the node one by one,resulting in inefficient communication.Nodes within the coverage of the receiver antenna can receive messages from the receiver,but they cannot directly reply to the receiver because they follow existing protocols.This dissertation considers the scenario where the receiver multicasts data to multiple nodes,and proposes a data feedback protocol based on link characteristic aware.The receiver selects the appropriate node as the master node according to the link characteristics and communicates with it normally.And the other nodes are slave nodes and listen to the communication process between the master node and the receiver.In this way,the proposed method can effectively reduce the node charging redundancy and improve the success rate of node data feedback,thus improving the node data distribution feedback efficiency.(3)Concurrent data transmission protocol for active nodes based on collision tolerance is proposed.The data collection in low-power wireless transmission follows the protocol of Carrier Sense Multiple Access.For reducing channel competition between nodes,the protocol provides that when the sender detects that the channel is occupied,the sender will conservatively choose the back-off mechanism to wait for a period of time.However,in the scenario of large-scale node deployment,if the nodes frequently compete for channels and adopt the back-off principle,the delay of sensing data transmission will be greatly increased.As a result,the system can not collect node sensing information in real time.Different from the traditional protocol of back-off mechanism,this dissertation proposes a protocol of concurrent data collection of multi-hop node based on collision tolerance.By adjusting the concurrent transmission probability of the competing nodes,the vicious competition between the nodes is reduced,so that they are in the competitive dividend interval.This method can effectively reduce the delay of data transmitting,and then it can improve the channel utilization of the system.(4)A system of behavior sensing for low-power wireless transmission is designed.Most of the existing behavior sensing systems use machine learning to analyze and process data.In order to obtain higher processing accuracy,a relatively large number of training sets are needed.However,the data training set is usually limited,which greatly affects the accuracy of the behavior sensing system based on the method of machine learning.A high-precision behavior sensing system based on High Order Cumulants is designed for the case of limited data training set.This work dose not only analyze the effectiveness of the proposed method in theory,but also verify the good performance of the system through a large number of experiments.At the same time,the system can be compatible with the existing commercial receivers,which has broad use value and practical significance.