软件定义传感网的网络重配置若干算法研究

Combining software defined network technology with wireless sensor network, the definition and architecture of software defined sensor network (SDSN) are proposed. The network reconfiguration technologies of SDSN including coverage optimization, topology control and routing optimization are researched in detail, in order to acquire the innovation achievements on the key technologies and related theories. The contents and objectives for this research are as follows. Firstly, the theory of Voronoi graph is used to search the optimal allocation of sensing radius to ensure the energy balance in full coverage, and also to find the optimal subset of sensor nodes and their sensing radius to achieve K-coverage on target region. Secondly, based on the theory of simplicial complex, the efficient calculating algorithm of edge chain group minimal generator is studied in the pre-power allocation, and the efficient iterative algorithm is also studied in the power reconfiguration. Finally, the optimal algorithm for routing is studied to guarantee the QoS requirement of each serve in the multi-serve transmission and to balance the residual energy of each sensor node in the whole SDSN. Through in-depth theoretical analysis, carefully designed computer simulation and necessary assistant experiments, the proposed algorithms will be verified. The proposal aims to systematically complete the above tasks, making valuable contributions to practical wireless communication systems.

本申请将软件定义网络技术与无线传感器网络相结合，提出软件定义传感网的架构，给出其定义，拟对其网络重配置技术进行系统深入地研究，包括覆盖优化、拓扑控制、路由优化机制等方面，力争在软件定义传感网相关理论和关键技术方面有所创新。具体研究内容及目标为：运用Voronoi图理论，寻求软件定义传感网全覆盖的最优感知半径分配方法以保证网络能量均衡，选择最优节点子集和其感知半径以达到目标区域的K-覆盖；基于单纯复形理论，研究软件定义传感网功率预配置中的边缘链群最小生成元高效、次优算法，以及快速、高效的软件定义传感网功率重配置迭代算法；研究基于多业务服务质量（QoS）的软件定义传感网路由优化机制，以保证多个业务同时进行数据传递和业务的QoS，并平衡传感器节点的剩余能量。本申请将通过深入的理论分析、计算机仿真并辅之必要的实验，验证所提算法的合理性，为将来能够应用于实际无线通信系统奠定基础。

本项目将软件定义网络技术与无线传感器网络相结合，提出了软件定义传感网的架构，给出其定义，并深入研究了“基于Voronoi图的动态覆盖算法”、“基于单复形理论的拓扑控制算法”和“基于多业务QoS的路由算法”等内容。具体包括：研究了运用Voronoi图理论，寻求软件定义传感网全覆盖问题中保证网络能量均衡的最优感知半径分配，以及选择最优节点子集和其感知半径，以达到目标区域的K-覆盖；基于代数拓扑的同调理论，研究通过计算软件定义传感网拓扑结构的最小边缘链群保证网络的基本连通性和覆盖需求，并进行有效节点删除的算法，以及基于移动节点辅助的网络拓扑结构重配置的迭代算法；研究基于多业务QoS的软件定义传感网路由优化机制，以平衡传感器节点的剩余能量为目标，保证多个业务同时以各自QoS进行数据传递，以及以最小能量路径为目标，研究链路干扰最小的能效路由算法，并进一步拓展到路径中存在障碍物的能效中继选择算法。通过以上诸点的研究，较好地解决了软件定义传感网中涵盖覆盖、拓扑和路由的关键性问题，为将来能够应用于实际无线通信系统奠定基础。