用户移动及网络拓扑自适应的可见光通信网架构及理论研究

This project will construct an adaptive framework of visible light communication (VLC) network for supporting user mobility and network topology in order to solve the problems of large interference among the different VLC cells, long handover delay, large outage rate and the rapid resource allocations that raised by the dynamic variations of user mobility and network topology. The details of the main research contents are as follows: (1) design the distributed quality of experience (QoE) coverage hole detection algorithm based on the homology theory and model the QoE flexible coverage to localize the QoE coverage holes in VLC network; (2) propose the user-centric based adaptive optimized multi-user transmission mechanism for supporting dynamic variations of network topology to solve the interference among different VLC cells; (3) study the sparse Bayesian learning theory for predicting the user mobility characteristic and propose the dynamic advanced handover algorithm based on fuzzy logic to reduce the handover delay; (4) for the resource allocation problem when the users are moving and the network topology is changing, the project will propose the VLC power optimization algorithm based on the minimization of QoE fluctuation and a new resource allocation algorithm based on the non-cooperation game theory to utilize the network resources efficiently.The expected research results will overcome the drawbacks of the invalid supports for the user mobility and the network topology variation in the VLC network. The research results will establish the foundations for the theoretical and technical developments of the future VLC high speed networks.

本课题拟构建用户移动及网络拓扑自适应的可见光通信网架构，以解决在可见光通信系统中由于用户移动和网络拓扑变化所带来的VLC小区间信号干扰大，移动切换时延长，传输中断率高以及资源快速分配等问题。具体内容为：（1）设计基于同调理论的分布式QoE覆盖空洞检测算法，构建QoE柔性覆盖区域模型，实现对可见光通信网QoE覆盖空洞的定位。（2）研究用户中心网络拓扑自适应优化的多用户传输机制，解决用户移动及网络拓扑动态变化下VLC小区之间的多用户干扰问题。（3）研究稀疏贝叶斯学习方法对用户移动特性进行预测，进而提出模糊控制的动态预切换算法，降低切换时延。（4）针对用户移动及拓扑动态变化时的网络资源优化分配问题，提出最小QoE波动的功率优化算法和非合作博弈资源分配算法，实现网络资源的高效利用。预期研究成果将克服传统可见光通信网无法支持用户移动及网络拓扑变化的缺点，为未来实现高速可见光通信网奠定理论与技术基础。

本课题构建用户移动及网络拓扑自适应的可见光通信网架构，以解决在可见光通信系统中由于用户移动和网络拓扑变化所带来的VLC小区间信号干扰大，移动切换时延长，传输中断率高以及资源快速分配等问题。本课题的主要成果为：（1）提出用户体验质量（QoE）概率覆盖模型，并基于该模型分析可见光NOMA接入及用户配对策略对室内可见光通信系统覆盖性能的影响；（2）提出基于切赫复形的LED覆盖空洞检测算法，并给出LED室内优化布局方案，实现可见光传输和照明的室内全覆盖；（3）设计多色 LED 可见光物联网架构，提出以用户为中心，联合优化QoE和室内照明质量的多用户调度算法。该算法不仅提高了用户平均 QoE，而且在确保照明质量的同时网络能服务更多的用户，极大地减少了小区间干扰；（4）提出一种基于最大化用户QoE的垂直切换算法，该算法将切换问题转化为马尔可夫决策问题，切换决策以增强QoE并降低用户的切换成本为依据。（5）提出LED簇间空洞与资源分配权衡算法，满足通信质量条件下最小化室内覆盖空洞。项目执行过程中发表了SCI检索刊物论文13篇，EI检索国际会议论文3篇，EI检索期刊论文3篇；授权国家发明专利8项，受理3项。项目培养了青年教师1名，博士研究生2名，硕士研究生6名。