无线多媒体传感器网络中基于网络编码的QoS感知传输控制技术研究

Wireless Multimedia Sensor Network (WMSN) is recently emerging as an extension to traditional scalar wireless sensor network, with the distinctive features of sensing rich multimedia content such as audio, images, video and requiring capability for massive data transmission. The multimedia streaming is sensitive to Quality of Service (QoS) parameters, such as delay, delay jitter, bandwidth and so on. How to provide QoS support for the multimedia streaming applications is the key to the practical application of WMSN. Research of QoS-aware WMSN transmission control protocol is developed by using the theory of network coding and fuzzy logic. Firstly, WMSN QoS multi-path routing protocol based on network coding and fuzzy logic is studied, and an energy-aware clustering algorithm based on fuzzy logic is designed, to form clustering network topology. A QoS routing selection algorithm based on fuzzy logic is proposed, which optimizes the multi-path routing selection. And the network coding based multi-path routing model is established, which is used to analyze the energy and transmission performance of WMSN. Secondly, QoS-aware multi-path congestion control mechanism is investigated by combining the theory of network coding, QoS multi-objective optimization and congestion control. To settle the congestion problem of multi-path routing, network coding-aware multi-path congestion control multi-objective optimization model is established, and the Grey-fuzzy theory is used to find the most optimum solution within multinomial time. Finally, service-aware congestion control scheme based on priority is studied. The fuzzy buffer management algorithm of storing coded packets, priority-based fuzzy traffic scheduling algorithm, and congestion aware multi-path fuzzy traffic allocation based on priority algorithm are proposed. By this program research, it could be expected to achieve greater breakthroughs in QoS transmission control protocols, and speed up the development and practical application of WMSN.

WMSN中多媒体信息对QoS性能指标敏感，如何保障多媒体信息的QoS传输是WMSN实用化的关键。项目基于网络编码和模糊智能控制理论，开展QoS感知的WMSN传输控制技术研究。首先,研究基于网络编码和模糊逻辑的WMSN QoS多路径路由协议，设计基于模糊逻辑的能量感知分簇算法，构建分层分簇网络结构；提出基于模糊逻辑的QoS选路算法，实现多路径路由选择的优化；建立基于网络编码的多径路由模型并分析其传输性能。其次,结合网络编码、QoS多目标优化与拥塞控制，建立网络编码感知的多路径拥塞控制多目标优化模型；利用灰色模糊理论，在多项式时间内求得最优解，解决编码网络多路径传输的拥塞问题。最后，研究服务感知的基于优先级的拥塞控制，提出模糊编码包缓冲区管理、模糊流调度和基于优先级的拥塞感知模糊多路径流量分配算法。通过项目研究，可望在WMSN的QoS传输控制技术方面取得突破，对推进WMSN实用化具有重要意义。

WMSN中多媒体信息对QoS性能指标敏感，如何保障多媒体信息的QoS传输是WMSN实用化的关键。项目基于网络编码和模糊智能控制理论，开展QoS感知的WMSN传输控制技术研究。首先，研究基于模糊逻辑的WMSN QoS多路径路由协议。一是基于本地信息采用模糊控制的分簇方法对网络进行划分，提出了一种基于簇的QoS多路径路由协议CQMRP，通过引入节点饱和度和路径饱和度的概念，有效避开拥塞节点，实现簇间具有QoS的节点不相交多路径路由发现。二是设计基于模糊逻辑的能量感知分簇算法，构建分层分簇网络结构；提出基于模糊逻辑的QoS选路算法，实现多路径路由选择的优化；并结合网络编码进行数据传输。 其次，结合网络编码、QoS多目标优化与拥塞控制，建立网络编码感知的多路径拥塞控制多目标优化模型；利用灰色模糊理论，在多项式时间内求得最优解，解决编码网络多路径传输的拥塞问题。并针对无线多媒体传感器网络（WMSN）中多对一通信时产生的网络拥塞问题，提出了一种自适应的WMSN网络拥塞控制算法ACCP。然后，针对WMSNs节点能耗问题，提出了一种基于粒子群优化的多路径自适应流量分配路由MATDR-PSO算法，通过建立能效优化模型，将最小能耗和能耗均衡2个特性转化为多目标优化问题；利用粒子群优化算法动态分配多路径上各条链路流量来实现网络的能效优化。最后，研究服务感知的基于优先级的拥塞控制，提出模糊编码包缓冲区管理、模糊流调度和基于优先级的拥塞感知模糊多路径流量分配算法。此外，项目进一步研究了考虑网络生存性的路由机制和MAC协议的设计。相关研究成果在WMSN的QoS传输控制技术方面取得了一定突破，对推进WMSN实用化具有重要意义。共发表学术论文28篇（含1篇录用），其中EI检索9篇，论文共计被他引14次。申报国家发明专利1项。项目组在该方向上培养硕士研究生1名，相关方向上培养博士研究生1名，硕士研究生2名。