Energy Efficient Routing Technology In Large-scale Wireless Sensor Networks

As the current application scenarios become more and more complex,the scale of wireless sensor network(WSN)is also increasing.An important feature of WSN is that multiple sensors work together and the types of sensing data are abundant.One of the obvious characteristics of data types are periodic data packets with strong spatio-temporal correlation,and the other is independent continuous data flow with large traffic.At present,the research on energy-efficient routing of WSN based on data types is not extensive.From the perspective of generating different data types in large-scale WSN,this thesis studies the corresponding energy-efficient routing protocols for scenarios with different data characteristics.The research focuses on energy-efficient routing mainly include: considering the energy-efficient routing of data aggregation from the scenario of strong spatio-temporal correlation,and considering the multi-path load balancing energyefficient routing from the scenario of continuous data flow.EEDADQN-routing algorithm is proposed for the scene with spatiotemporal correlation of data in large-scale WSN.Firstly,the scene is analyzed and described,and the scene characteristics and requirements are displayed.According to the analyzed characteristics and requirements,the dynamic sending queue model and data aggregation degree model are established.These two models are applied to the DQN algorithm.For the spatio-temporal correlation scene of WSN,DQN is used to model the routing algorithm to form the DQN routing algorithm EEDADQN routing based on data fusion.Finally,the algorithm is implemented and simulated in ns-3 and ns3-gym platforms.Compared with other similar algorithms,the algorithm proposed in this thesis has more advantages in network lifetime and average hops.FZ-ELQAOMDV algorithm is proposed for the scenario of continuous data flow in large-scale WSN.Firstly,the scene characteristics are analyzed,and then the shortcomings of typical multipath routing protocol AOMDV are analyzed.and the AOMDV is improved according to its shortcomings and scene characteristics.In the aspect of routing selection,the RREQ message is modified,and the reverse routing algorithm is improved,and then the RREP message is improved to collect path information.The path information is re-evaluated by fuzzy logic at the source node,and finally the path adaptation proposed in this thesis is used.value for flow distribution,so far the AOMDV transformation is completed.Finally,the algorithm is implemented and simulated on the ns-3 platform.Compared with other similar algorithms,it has more advantages in network delay and network energy efficiency.