Research On Routing In Software Defined Internet Of Things

The Internet of things(IoT) and software defined network (SDN are the core technologies of new generation of information and network. Especially, SDN is a hot research topic in computer networks nowadays. In the era of IoT, the physical infrastructure of heterogeneous network system will be more complicated. Thus, the introduction of SDN for IoT can provide a mechanism for the flexible, agile and personalized management of sensing, networking and transmissions. As well as facilitating the core cellular network to connect with edge devices or sensors nodes in an energy-efficient fashion. IoT will also become the main driving force for the development of SDN in terms of promising and intelligent applications and services, such as smart city, urban computing and ubiquitous healthcare monitoring.From the aspect of IoT sensing, wireless sensor network is the basis of the Internet of things. Routing in wireless sensor networks has been considered as an important research area for many years. Currently, software defined networking in the field of WSNs is at the beginning and exists various challenging issues. The alternative methods of software defined routing to traditional routing are still in the study. In fact, the idea of software defined network has been applied into some traditional routing algorithms of IoT.In order to promote the technologies of IoT routing and SDN more effectively, this dissertation investigates the integration of SDN and IoT and the architecture of software defined Internet of things. The challenges and enabling technologies of routing in software defined IoT are studied. The proposed schemes are verified through extensive network simulation by OPNET.In the cooperative communication and intelligent agent, a novel cooperative communication algorithm of IoT to construct the architecture of cooperative communications, and data are forwarded by K-cooperative nodes (KCN) per hop. Idle nodes go to sleep periodicity to save energy. The KCN algorithm can maintain high transmission reliability in highly dynamic wireless sensor networks; We have also studied the problem of route planning assisted by multi-agent for parallel access to multiple source nodes. The directional source grouping based multiple intelligent agent algorithm is proposed, where the length of route is directionally controlled by the angle of sector.For IoT multimedia routing, the multimedia disjoint construction mechanism multipath planning (M-Plan) is proposed, which includes multipath planning for single source (MPSS) scheme and multipath planning for multiple source scheme (MPMS). Multiple paths for concurrent transmissions are built according to B-Splines curves. The minimum energy consumption could be achieved while the user requirement of QoS delay can be guaranteed. We apply the idea of SDN into M-Plan to enhance the capability for multiple source routing coordination strategy. The routes of source nodes are programmed and controlled dynamically. The algorithm of pair-wise directional geographical routing (PWDGR) is put forward to alleviate the problem of energy hole of the nodes near Sink, PWDGR can balance the load of data in the vicinity of the Sink node to extend the life of network.In realization of software defined IoT method, this thesis presents a new hardware-in-the-loop (HIL) emulation method to emulate the SDN. We use the System-In-The-Loop (SITL) technology of OPNET to build the SDN HIL emulation platform of healthy network. The emulation test platform includes the sensor network part or all of which may use the software simulation, the transmission network and the cloud data center both of which use the real network. It can fully save the energy consumption under the condition of assuring the demand of QoS application.