云环境下基于文化基因算法的大规模异构无线传感器网络节能覆盖控制方法研究

The coverage is the monitor standards of the Quality of Service(QoS) and application foundation for Wireless Sensor Networks(WSNs). Deployment of sensor nodes with large-scale and high-density, heterogeneous sensing and communication resource, energy constraints make the coverage control problem in WSNs not only a NP-C Mixed-Integer Nonlinear Programs (MINLPs), but also a large-scale, high-dimensional optimization problem. There's no evidence that any effective method can optimize the subset of working node, the sensing and communication resource all at once. Our project aims to improve the optimization model for coverage control problem of WSNs and set up a new model to optimize the subset of working node, the sensing and communication resource with the goal of minimizing energy consumption and on the condition of coverage and connectivity. We propose the large-scale optimization algorithm under the framework of memetic, using global search and local search algorithms through knowledge rules guide, so that the algorithm has faster convergence speed and precision. We will study the Map-Reduce programming model under the cloud platform, and modify its adaptability to improve the speed ratio. Through the research of our project, it is expected to construct the basic theoretical framework for large-scale MINLPs with high dimension and distributed processing mode in theory. We will also explore the adaptive balance mechanism between global search and local search of memetic algorithm. And we expect to provide the energy-efficient coverage control scheme in the practical application of large-scale heterogeneous wireless sensor network.

传感器节点大规模高密度部署、异构感知和通信资源、能量受限使WSNs覆盖控制问题不仅是一个NP-C混合整数非线性规划问题(MINLPs)，而且具有高维和大规模的特征，目前尚未找到同时优化工作节点子集、感知和通信资源的有效方法。本项目拟改进WSNs的覆盖控制模型，以最小能耗为目标，以连通覆盖为条件，建立同时优化节点工作子集及感知和通信资源的问题模型；研究基于文化基因算法框架的大规模优化算法，利用知识规则对算法的全局搜索和局部搜索进行引导，从而使算法具有更快的收敛速度和求解精度；研究云平台下的MapReduce 编程模型，并对其适应性修改，提高加速比。通过本项目的研究，可望在理论上建构高维大规模混合整数优化问题的基础理论框架和分布式处理模式，探索文化基因算法的全局和局部搜索自适应平衡机制；在实践上为大规模异构WSNs的应用提供节能覆盖优化方案。

本研究以大规模无线传感器网络为研究对象，结合其在部署时对节能的需求，从其覆盖控制的角度出发，综合异构传感器节点的感知模型和通信模型的特点，在保证覆盖和连通要求的前提下，以整个网络的最小能耗为目标，从大量冗余的传感器节点中选择最优工作子集。本项目在执行过程中，以设定的研究目标为导向，完成了在对感知区域完全覆盖的前提下，考虑了不同类型传感器节点的通信能力、工作负载、运营成本等影响WSN网络能耗的因素，将能耗转化为部署成本，提出了在降低部署成本的同时，尽可能保证探测质量的有效方法。以最小化所需传感器节点的数量和所有传感器节点的功能冗余为目标，研究了无线传感器网络中多功能传感器的配置问题。以可靠的通信质量和合理的成本控制，作为移动传感器网络中Sink节点布置问题的两个主要目标，研究了移动传感器网络Sink节点布置优化算法。特别是对于大规模无线传感器网络的传感器放置问题，提出了一种基于Spark的两阶段遗传算法（SGA），SGA在准确性和效率上均优于其他传统算法。本项目的理论研究成果对提高和优化大规模异构无线传感器网络的覆盖质量和延长网络运行寿命具有直接的作用。以此为基础，能为异构无线传感器网络应用于各类资源环境监测及其它应用场景提供理论支撑及原型参考。