低占空比无线传感网络广播与机会路由协议的研究

发布时间：2018-05-12 07:52

本文选题：无线传感器网络 + 低占空比　；参考：《中国科学技术大学》2017年博士论文

【摘要】：大量的无线传感网应用需要将数据包从基站分发到网络中的每个传感器节点或者将传感器节点收集到的信息传递到基站节点,因此,需要设计高效的广播与路由协议。由于低占空比睡眠调度与不可靠无线传输的影响,使得无线传感网中的广播与路由协议不同于其他类型网络中的协议。无线传感网中传统的广播协议存在广播进度信息收集不完整、对广播过程的动态性适应能力差以及对于侦听到的信息利用不充分等问题。传统的机会路由策略中忽略了节点独立睡眠调度为网络带来的动态性,当发送节点面对不同的邻居节点唤醒次序时,需要对路由策略进行动态调整。本文重点研究了如何处理网络中动态变化的因素,为低占空比无线传感网设计出高效、快速的广播与机会路由协议。主要内容与创新点包括:1.提出了考虑强制发送方之间合作的基于状态的高效广播协议:传统的基于状态的广播由于潜在的发送节点间低效的合作机制引入大量的冲突与冗余传输,降低了广播的效率。本文基于提出的信息编码与解析方案实现高效的侦听机制,节点能够获得更加准确的广播进度信息,从而在不引入额外负载的前提下使得发送节点之间能更好的合作,提升广播的速度与效率。提出的协议能够在不引入额外时延的条件下解决稀疏网络中的四边形拓扑问题,并且能够依据应用需求对速度与效率进行权衡。实验结果表明提出协议的性能优于传统的基于状态的广播协议。2.提出了基于网络编码与调度生成树的广播协议:在基于调度生成树的广播中,树中拥有相同父亲节点的节点同时被唤醒,从而利用无线信道的广播特性。节点间交互网络编码包数信息,基于实时的包数信息,接收节点决定哪些邻居节点能够提供额外的编码包。为了保证低占空比,接收节点限制可侦听节点的数目。提出一个多项式级别时间复杂度的动态规划算法确定一个节点数目固定的最优侦听节点集合,从而最大化一个睡眠调度周期内期望收到的编码包数。仿真结果表明提出的协议能够有效的降低传统广播协议的时延与能耗。3.提出了基于网络编码与移动基站节点的广播协议:传统广播机制通常由一个静止的基站作为广播源。基于移动基站,当其移动到某个位置,可以作为新的广播源。多个广播源同时转发数据包,可以极大提升广播速度。本文将从源节点传输多个编码包到某个节点的过程建模成一个Markov过程,分析并估计从广播源到节点的期望时延。广播树被划分成多个分支,基站可以沿着某条分支移动或者在分支间转移。考虑所有的分支访问顺序后,为基站确定一个最优的能够最小化广播时延的移动路径。仿真结果表明在移动基站的协助下可以显著的降低广播时延。4.提出了考虑独立睡眠调度的最小化时延的机会路由协议:在机会路由中,发送节点依据实时的网络状况动态的从候选节点集合中选择转发节点。由于独立睡眠调度的影响,当发送节点在不同时刻发送数据包时,有可能面对不同的候选节点唤醒序列。传统的机会路由协议忽略了候选节点唤醒序列的时变特性。本文研究了唤醒序列的动态特性,从理论上分析如何为节点确定能够最小化端到端时延的候选节点集合。考虑到机会路由协议实际部署时面临的挑战,提出了全局优化与局部优化相结合的机会路由策略。依据相对静态的全局信息(如网络拓扑与节点调度周期信息)建立初始候选节点集合,然后在局部逐跳考虑动态因素(如实时的链路质量与节点唤醒序列)的影响。实验结果表明提出的算法能显著的提升机会路由的性能。
[Abstract]:A large number of wireless sensor networks need to distribute packets from the base station to each sensor node in the network or transmit the information collected by the sensor nodes to the base station nodes. Therefore, it is necessary to design efficient broadcast and routing protocols. The wireless sensor network is made due to the influence of low duty cycle sleep scheduling and unreliable wireless transmission. The broadcast and routing protocols in the radio and routing protocols are different from those in other types of networks. The traditional broadcasting protocol in wireless sensor networks has the problems of incomplete broadcast progress information collection, poor dynamic adaptability to the broadcast process and insufficient use of the detected information. The traditional opportunity routing strategy ignores the independent sleep of nodes. It is necessary to adjust the routing strategy dynamically when the sending node faces different neighbor nodes. This paper focuses on how to deal with the dynamic changes in the network and design a high efficient, fast broadcast and opportunity routing protocol for the low duty wireless sensor network. The innovation points include: 1. proposed a state based efficient broadcasting protocol considering the cooperation between the forced sender: the traditional state based broadcasting has reduced the efficiency of broadcasting due to the introduction of a large number of conflicting and redundant transmission due to the inefficient cooperative mechanism between the potential sending nodes. The effective interception mechanism allows the nodes to obtain more accurate broadcast progress information, thus making the sending nodes better cooperative and improving the speed and efficiency of the broadcast without introducing additional loads. The proposed protocol can solve the quadrangular topology problem in sparse networks without introducing additional delay, and can also solve the problem of the quadrangular topology in the sparse network. The experimental results show that the performance of the proposed protocol is better than that of the traditional state based broadcast protocol (.2.). The broadcast protocol based on the network coding and scheduling spanning tree is proposed. In the broadcast based on the scheduling spanning tree, the node with the same father node is awakened at the same time. The broadcast characteristics of a wireless channel. The number of packets between the nodes and the number of packets, based on the real-time packet number information, the receiving node determines which neighbor nodes can provide additional coding packets. In order to ensure the low duty ratio, the receiving node limits the number of the nodes to be listener. A dynamic programming algorithm with multiple level time complexity is proposed. Determine a set of optimal number of nodes with fixed number of nodes to maximize the number of coded packets expected in a sleep scheduling cycle. The simulation results show that the proposed protocol can effectively reduce the time delay and energy consumption of traditional broadcast protocols..3. proposed a broadcast protocol based on network coding and mobile base station nodes: traditional broadcasting. A mechanism usually consists of a stationary base station as a broadcast source. Based on a mobile base station, when it moves to a certain location, it can be used as a new broadcast source. Multiple broadcast sources can forward packets at the same time, which can greatly improve the broadcast speed. This paper will model a Markov process from a source node to a node. The expected delay from the broadcast source to the node is estimated. The broadcast tree is divided into multiple branches, the base station can move along a branch or transfer between the branches. After considering all the branch access order, the base station can determine an optimal mobile path to minimize the broadcast delay. A significant reduction in broadcast delay.4. presents an opportunistic routing protocol that considers the minimum delay of independent sleep scheduling: in an opportunistic routing, the sending node selects the forwarding node from the candidate node set according to the real-time network status. Due to the influence of the independent sleep scheduling, when the sending node sends packets at different times, the node has a data packet. It is possible to face different candidate node wake-up sequences. The traditional opportunistic routing protocol ignores the time-varying characteristics of the candidate node wake-up sequence. This paper studies the dynamic characteristics of the wake-up sequence and theoretically analyzes how to determine the candidate node convergence to minimize the end to end delay for the nodes. The opportunity routing strategy combined with global optimization and local optimization is proposed. The initial candidate node set is established based on the relative static global information (such as network topology and node scheduling cycle information), and then the effect of dynamic factors (such as real-time link quality and node wake-up sequence) is considered in the local hop by hop. It shows that the proposed algorithm can significantly improve the performance of opportunistic routing.

【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【相似文献】