The Research Of Three-dimensional Clustering And Dynamic Evolution Mechanism For Wireless Sensor Network

Wireless sensor networks (WSNs) is a burgeoning next-generation sensor network, it isdeemed as one of the most influential technologies which could change the world in the21stcentury, and it also is one of the key bottom technologies of Internet of Things. With the rapiddevelopment of the microelectronics technology, computer technology and wirelesscommunication technology，the nodes in the WSN have a fast development to miniaturization,low power consumption and multi-function. This makes the WSNs have a wide applicationapace，such as the applications in the field of military defense, environmental monitoring,public health and so on. However, as the breadth and depth of these applications growing,some new problems have come out, such as the network structure needs to extend fromtwo-dimensional flat to three-dimensional space, the serious energy dissipation, the unbalanceof residual energy, poor organization and anti-destroying ability and so on.At the same time, the complex networks theory as an important research tools is used invarious fields, and attracts more and more researchers’ focus. This article also uses thecomplex networks to simulate the dynamic evolutionary process of the WSN’s topologicalstructure, and analyzes the features of them for exploring a new way to solve the seriesproblems of the large-scale WSNs.The main research content is as follows:(1) Energy and distance adaptive clustering algorithm for3D wireless sensor networksBased on the practical application for the tunnel safety monitoring and combining withthe classic LEACH algorithm, we propose the energy and distance adaptive clusteringalgorithm for3D wireless sensor networks (3D-EDAC). This algorithm considers the residualenergy and the distance from node to Base Station, and seeks the optinal number of the cluterheads. Besides, improves the clutering process and the communication mode. Through thesimulation experiments, the3D-EDAC algorithm can adapt to the3D space well and prolongthe survival time of the network. The energy-saving effect is remarkable.(2) A Local-world Evolving Model for WSNs with the Self-regulating AttractivenessDue to the limit of node’s energy and communication, we introduct the local-world evolvingmodel into the WSN and reference the node’s residual energy and the distance between them,then, form a self-regulating attractiveness to implement the preferential attachment for thenew coming node. The experiments show that this model can reduce the connection degree ofthe key nodes, avoid the key nodes costing the excessive energy to maintain a larger numberof connections, so as to keep the nodes having a long survival time, and improve the invulnerability of the network.(3) A Strength-Constrained Weighted Evolving Model for Wireless Sensor NetworksCombining the weighted network into WSN, we simulate the data flow between thenodes, the key nodes and the data path. Based on the BBV evolving model, we propose astrength-constrained weighted evolving model for WSNs (SCW model). This model not onlykeeps the dynamic growth of the weights, but also limits the strength to avoid the nodestrength unlimited growth with the evolution of network. The results of simulationexperiments bring into correspondence with this model. Therefore, this model could simulatethe change of the connectivity among the nodes, and protect the network congestion causedby exorbitant strength and fast energy consumption, so as to improve the survival time of thenetwork.