层次型无线异构传感器网络资源重构与鲁棒路由联合设计

Within the constraint of multiple domain resources, improving the reliability of data transmission in wireless sensor networks (WSNs) and ensuring the quality of service has become key problems that trouble the WSNs researchers. The project will focus on the robust routing problem among heterogeneous nodes with hierarchical network topology, and correspondingly propose a robust routing strategy. Firstly, according to the feature properties of nodes, it will study the method of creating network topology, achieve the optimal matches between nodes with different functions, and then offer the network topology to member nodes. Next, in the domain of frequency, we will explore the spectrum detection method, evaluate the link state, and reformulate the network resources by expanding the narrow bands over wide spectrums. Based on this, the study results will provide a basis for spectrum sensing when establishing the feasible data routes with strong robustness. Furthermore, as considering the uncertain mathematical characterization of session rate, in compliance with the required evaluation index system the project will use the portfolio investment in economic mathematics to establish reliable data routes and the link rate allocation scheme, which is immune to the parameter perturbation. The research work of the project will realize the deepening application of cognitive technology in routing problems of WSNs, and will provide both the solid theory basis and the technical support for reliable data transmission in WSNs.

在多域有限资源约束下，提高传感器网络数据传输的可靠性，保证网络服务质量，成为日益困扰WSN研究者的关键问题。本项目关注层次型网络拓扑中属性异构节点间的路由鲁棒性问题，并构建具有鲁棒性保证的路由策略。首先，根据认知节点的功能属性，研究网络拓扑创建方法，实现节点间优化匹配，为成员节点的路径选择提供拓扑基础。其次在频域范围内，研究频谱检测与链路状态评估方法和"窄带宽频"的资源重构手段，为建立具有鲁棒性保证的可行数据路径提供感知依据。进一步的，当考虑节点间会话速率具有不确定的数学特征时，综合运用经济数学中的组合投资策略，依据网络所需指标体系，在多影响因素下建立可靠的数据路径和参数摄动免疫的链路速率分配方案。本项目的研究将进一步深化认知技术在无线传感器网络路由构建中的应用，并可为提高传感网中数据传输的可靠性提供理论依据和技术支撑。

该项目名称为层次型无线异构传感器网络资源重构与鲁棒路由联合设计。.首先，为满足覆盖与节能应用要求，提出了无线传感器网络中一种最小刚性拓扑控制算法。该算法基于Voronoi覆盖机制，准确控制节点工作状态，除实现活动节点对目标区域的完全覆盖，还利用Delaunay三角剖分图的特点，有效约束了网络平均节点度。其次在频域范围内，对异构无线网络中拓扑控制和网络资源分配问题联合设计，得到最优的分簇结果和资源分配方案，为建立具有鲁棒性保证的可行数据路径提供感知依据。最后，使用经济数学中的组合投资策略，联合网络寿命和鲁棒路由设计，引入改进受约束粒子群优化方法，求解非凸规划问题，得到资源配置的最佳方案。.本项目的研究将进一步深化认知技术在无线传感器网络路由构建中的应用，并可为提高传感网中数据传输的可靠性提供理论依据和技术支撑。