面向节能的多源异构无线传感网拓扑控制理论与算法研究

Energy constraint is one of the key problems in wireless sensor networks (WSNs). The differences of multi-source sensor nodes and different harsh environment which faces will exacerbate the complexity of the actual network model. However, topology control is an efficient way to solve such problems. In this project, it starts from optimization theory and graph theory while topology control theory and algorithms are proposed. Compared with the ideal network model, we will combine constraints from multi-source heterogeneous WSNs such as energy, time-delay, packet loss and clock synchronization to establish complex and effective mathematical model. Then some heuristic algorithms are analyzed and designed to obtain the optimal network topology using convex optimization theory. Furthermore, WSNs is described by graph. Combined with graph theory and data fusion algorithm, it needs to find a suitable and adaptive trigger condition to build connectivity-oriented and Qos preservation topology control strategy through the analysis of topology maintenance trigger mechanism. Considering the probability network model, we analyze and compute the network reachability based on reliability theory, and further proposes opportunity-based topology control algorithm. The research result of this project will make innovations and breakthroughs in topology control theory for energy-efficient multi-source heterogeneous wireless sensor networks. It also promotes the application in the field of WSNs and the Internet of Things industry, which further promotes its development and improvement.

能量约束是无线传感器网络(WSNs)最核心问题之一，多源传感器节点的差异性及面对不同恶劣环境都会加剧实际网络模型的复杂性，而拓扑控制是解决此类问题的一种有效手段。本项目从最优化理论和图论角度出发，提出拓扑控制理论与算法。相比现有理想的网络模型，拟结合多源异构WSNs 的能量、时延、丢包、时钟不同步等约束条件，建立复杂有效的数学模型，利用凸优化理论分析并设计启发式算法求解最优网络拓扑结构。将WSNs以图的方式描述，利用图论方法，结合数据融合算法，通过分析拓扑维护的触发机制，寻找合适的自适应触发条件，构建面向连通性和Qos保障的拓扑控制策略。考虑概率网络模型，基于可靠性理论分析计算网络可达性，进而提出机会拓扑控制算法。本项目的研究结果将在面向节能的多源异构无线传感器网络的拓扑控制理论上取得创新和突破，促进该理论及算法在无线传感器网络领域和物联网产业中的应用，并进一步推动其发展和完善。

能量约束是无线传感器网络(WSNs)最核心问题之一，多源传感器节点的差异性及面对不同恶劣环境都会加剧实际网络模型的复杂性，而拓扑控制是解决此类问题的一种有效手段。本项目从最优化理论和图论角度出发，提出拓扑控制理论与算法。引入最小连通支配集模型，构建节能的虚拟骨干树，设计了一种基于较小通信开销的网络拓扑，并针对负载不均衡问题，设计了一种虚拟骨干节点能耗收集协议。引入一种博弈思想，结合多源异构WSNs 的能量、传输时延、丢包、时钟不同步等约束条件，建立复杂有效的数学模型，从而建立基于势博弈的自适应拓扑控制方法，有效地提高网络的服务质量；进一步，将拓扑控制方法推广到水下异构无线传感器网络。综合图论、近邻理论、博弈理论等方法，分别考虑基于概率分布理论的树型拓扑和基于能量预测的分簇-树型拓扑，建立了一套完善的拓扑控制理论，进一步改善了拓扑控制在节能、传输效率与服务质量方面上的性能。最终，课题组共发表学术论文12篇，其中SCI收录论文4篇，EI收录论文6篇；申请发明专利13项，其中授权3项；授权软件著作权4项；同时设计了简易的拓扑控制算法仿真软件和开发了基于水下无线传感器网络(UWSNs)的多参数水质在线实时监测原型系统。本项目不仅力求在面向节能的多源异构无线传感器网络的拓扑控制理论上取得创新和突破，更积极探寻其研究成果在环境监测领域的应用，从而推动新兴物联网产业长足发展。