无线传感器网络视频流路由算法研究

发布时间：2018-03-11 14:45

本文选题：无线多媒体传感器网络　切入点：视频流数据　出处：《中国矿业大学》2017年硕士论文　论文类型：学位论文

【摘要】：无线传感器网络是目前热点研究领域。其中,由视觉传感器节点组成的无线多媒体传感器网络能够实现视频流的传输,在视频监控、环境监测、多媒体信息等领域有着广泛的应用。相比于传统的无线传感器网络,视频流数据具有数据量大、业务种类丰富、传输可靠等高要求。因此,对资源受限的无线多媒体传感器网络,研究保障视频流服务质量的方法是一个亟待解决的问题。为此,本文从路由算法和数据传输两个方面进行了尝试研究,主要内容如下:(1)针对无线多媒体传感器网络中的视频流传输需要提供多参数Qo S保障问题,本文提出一种基于蚁群优化的多路径Qo S路由算法AMRQR(ant-colony based multi-path Qo S routing)。算法综合考虑了网络的带宽、平均时延、丢包率等Qo S参数,将节点平均剩余能量作为路由选择的影响因素,对网络节点剩余能量进行分级,根据节点剩余能量等级动态调整能耗权重因子。同时,为了避免单路径传输所导致的网络负载不均衡,采用多路径传输的方式。仿真结果表明该算法在满足Qo S参数约束条件下,能够均衡网络能量,提高网络传输的实时性和可靠性。(2)针对无线传感器节点存储转发能力有限的问题,本文对无线多媒体传感器网络中视频监控图像序列进行研究。根据视频监控特点,利用改进的背景帧差分算法,对视频帧中的目标区域进行定位,仅传输目标区域。通过周期性的建立视频帧背景模型,对视频帧与背景帧进行帧间差分操作,根据两帧的差值大小,对传输的视频帧数据进行分级,进而根据不同等级选取相应的传输策略。实验结果表明,在保证传输图像质量的前提下,该方法能够有效的减少视频流数据传输量,大大降低传输能耗,均衡网络能量,延长网络生命周期。
[Abstract]:Wireless sensor network is a hot research field at present. Among them, wireless multimedia sensor network is composed of visual sensor nodes can realize video transmission in video surveillance, environmental monitoring, multimedia information, is widely used in many fields. Compared to the traditional wireless sensor network, streaming video data with a large amount of data, the type of business rich, reliable transmission of high requirements. Therefore, wireless multimedia sensor networks of limited resources, methods of security video streaming service quality is an urgent problem to be solved. Therefore, we try to study from the two aspects of routing and data transmission, the main contents are as follows: (1) for wireless multimedia sensor networks in the video stream transmission need to provide multi parameter Qo S security problem, this paper proposes a AMRQR Qo multi path routing algorithm based on ant colony optimization S (ant-colony Ba Sed multi-path Qo S routing) algorithm. Considering the network bandwidth, delay, packet loss rate Qo S parameters, the average node residual energy as the factors influencing the routing and classification of network node residual energy, according to the node residual energy level dynamic adjustment of energy weight factor. At the same time, in order to avoid the single path transmission the network load is not balanced, the multi path transmission mode. The simulation results show that the algorithm meets the Qo S constraints, can balance the network energy, improve the real-time and reliability of network transmission. (2) for wireless sensor nodes store and forward problem of limited ability, this paper focuses on the research of wireless video multimedia sensor network monitoring. According to the characteristics of video surveillance image sequences, using the background frame difference algorithm is improved, the localization of the target area in the video frame, only. Lose the target area. Through the establishment of the video background model for periodic, inter frame differential operation of video frame and background frame, according to the difference between the size of two frames, video frame on the transmission of data, and then select the corresponding strategies according to different levels of transmission. The experimental results indicate that under the premise of ensuring the quality of image transmission next, the method can reduce the amount of data transmission of video stream effectively, greatly reduce the transmission energy consumption, balance the network energy and prolong the network life cycle.

【学位授予单位】：中国矿业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【参考文献】