| Wireless sensor network has the characteristics of flexible networking,low price,convenient and efficient.It usually consists of a base station and thousands of wireless sensor nodes.Wireless nodes are usually deployed in harsh environment which makes the battery cannot be recharged and the network lifetime is limited.Therefore,extending the network lifetime is of primary importance.In current studies,it mainly designs the protocols in physical layer,data link layer or network layer of the network to improve the network performance.In the study of network layer,cluster routing scheme is a popular solution.However,in these schemes,the energy consumption of the sensor node which closes to the base station is too fast.It causes the energy consumption unbalanced of the network.This is the "hotspot problem".There are many researches on "hotspot problem" in the domestic and overseas,and there are two main solutions.The first one is the unequal cluster-based routing schemes and the second one is the mobile base station-based routing schemes.For cluster routing scheme,the network will reconfigure and deploy the whole network routing in each round.This mechanism balances the network consumption,but it increases the control overhead and affects the link quality.Software-defined wireless sensor network separates the control plane and the data plane.It reduces the computation overhead and the control overhead,which improves the network performance.At present,there are still some problems in the research of the routing protocol in wireless sensor network.Firstly,when setup the routing in traditional wireless sensor networks,the control overhead and the packet loss rate is high.The network computation overhead is high.For low-cost network devices,the complex protocols cannot be deployed.Secondly,the clustering scheme is seldom used in the software-defined wireless sensor network.Because it doesn’t have the data aggregation process,network energy consumption is high.Thirdly,the cluster routing scheme exists the "hotspot problem"which affects network lifetime.The unequal cluster routing scheme needs to reasonably design the inter-cluster routing,in order to balance the network energy consumption.Fourthly,the routing frequent changes in the cluster routing-based scheme.It increases the number of control messages,which affects the stability of the network.According to the above analysis,the research of this dissertation mainly focuses on the following three parts.(1)Fuzzy-based multi-hop unequal cluster routing schemeThis part proposes the fuzzy-based multi-hop unequal cluster routing scheme.The scheme consists of the cluster head election phase,cluster formation phase,multi-hop routing phase and data transmission phase.In the cluster head election phase and the cluster formation phase,the scheme uses the fuzzy logic-based unequal cluster algorithm.Meanwhile,it proposes the parameters of "relative inter-cluster overhead" and "relative intra-cluster overhead".In the unequal cluster process.The input parameters of the fuzzy logic system are "relative residual energy","relative inter-cluster overhead"and "relative intra-cluster overhead".This method balances the energy consumption between the cluster heads and the cluster members.In the multi-hop routing phase,it uses the inter-cluster relay routing communication algorithm.It also proposes the parameters of "relative cluster relay overhead".Two factors are considered in the designing of the inter-cluster routing,which reduces the energy consumption of cluster heads.These two factors are the "relative residual energy" and the "relative cluster relay overhead" of the cluster head.A selfadaptive cluster head rotation mechanism is proposed in the data transmission phase.It reduces the probability of cluster recalculated and thus reduces the number of control message in the entire network.According to the simulation results,the proposed scheme balances and reduces the energy consumption,meanwhile prolongs the lifetime of the whole network.(2)Software-defined and fuzzy-based multi-hop unequal cluster routing schemeThis part proposes a software-defined and fuzzy-based multi-hop unequal cluster routing scheme.The scheme consists of the topology discovery phase,flow table generation phase,flow table delivery phase as well as data transmission and topology maintenance phase.It uses three methodologies to improve the performance of the routing scheme.Firstly,it designs a novel software-defined wireless sensor network architecture which is used in the static-based wireless sensor network.The network consists of a controller and many sensor nodes.The sensor node only needs to transmit its location and residual energy information to the controller.The controller is responsible for calculating and delivering the routing of the whole network.It reduces the computation and routing overhead of sensor nodes.Then,the scheme designs the controller-based fuzzy multi-hop unequal cluster algorithm.The algorithm first uses the fuzzy logic system generating the routing between cluster heads and cluster members.Then,it uses the fuzzy logic system generating the multi-hop routing between cluster heads.The algorithm makes the energy balance among all the sensor nodes.Finally,a residual energy-based rotation mechanism is proposed to reduce the control overhead.Simulation results show that this scheme can balance and reduce the energy consumption of the network,prolong the network lifetime,reduce the control overhead of the network.(3)Particle swarm optimization and artificial bee colony algorithm for clustering and mobile based software-defined wireless sensor networks schemeThis part proposes the particle swarm optimization and artificial bee colony algorithm for clustering and mobile based softwaredefined wireless sensor networks.It aims at designing a mobile base station-based cluster routing scheme.Balancing the network energy consumption and prolonging the network lifetime are the purposes of the scheme.In order to reduce the energy overhead and computation overhead of sensor nodes,a mobile base station-combined softwaredefined wireless sensor network architecture is designed.A particle swarm optimization-based cluster routing algorithm is used to generate the set of the cluster heads and the sojourn locations of the base station.The algorithm proposes the parameters of "the relative transmission overhead","the relative residual energy","the discrete degree of the cluster members in each cluster" and "the discrete degree of the sojourn locations for base station".These four factors are considered comprehensively.It reduces the energy consumption and indirectly reduces the network latency.The artificial bee colonybased traversal path algorithm is used to designing the optimum moving path of the base station.The two heuristic algorithms which are used in the proposed scheme are suitable for solving the continuity problems.By improving these two algorithms,the heuristic algorithms are suitable for solving the discrete problems of the proposed scheme.Comparing with other schemes,the proposed scheme reduces the energy consumption,prolongs the network lifetime and reduces the control overhead. |
