无线Mesh网络中面向网络编码的调度机制研究

发布时间：2018-08-23 07:52

【摘要】：无线Mesh网络由一组具有动态组网能力且无需基础设施支撑的可移动终端组成。与Ad hoc网络相比,无线Mesh网络由于具有大容量、高速率、低成本以及可扩展性好等优点,近年来得到工业界和学术界的广泛关注,并成为商业化“最后一英里”无线宽带热门接入技术。随着物联网的发展和无线频谱资源需求的日益增加,无线Mesh网络被广泛应用于城域网无线接入、车载通信和智能终端等方面。随着大数据网络通信时代的到来,无线多媒体业务正取代传统的语音和数据通信业务成为网络业务的主流。如何满足用户日益增长的带宽需求是研究者需要考虑的重要课题。尽管近年来相关学者提出多种提高无线Mesh网络吞吐量的技术方案,但是现有研究方案对不同技术方案应用时相互作用(如网络编码和空分复用调度,功率控制和网络编码)的研究较少。此外,现有基于无线Mesh网络调度的研究大都假设网络信道状态信息是可预先获知的,然而,实际无线网络信道环境实时变化,信道状态较难预测。同时,随着物联网相关研究如火如荼地进行,移动终端将越来越多地反映人的社会属性,这对新兴社交网络环境下的通信研究提出了挑战。本文针对上述问题和挑战,分别从基于信道增益和网络编码的调度机制、马尔科夫模型下基于网络编码的调度机制、多射频多信道多速率网络编码感知的调度机制、而向社交网络和网络编码的调度机制四方面对无线Mesh网络中面向网络编码的调度机制进行研究。本文的主要工作和取得的创新成果如下：(1)由于无线信道环境实时变化,信道状态较难预测,第2章提出基于无线信道增益和网络编码的机会调度方案和功率分配方案。本章首先基于模拟网络编码(Analog Network Coding, ANC)和时分广播(Time Division BroadCasting, TDBC)的网络模型提出一种采用滑动采样窗口的中断概率闭式解表达式,进而对无线信道增益进行估计。本章提出的信道增益估计策略能够适应网络动态变化,同时相比基于底层信息采集的方案,具有简单易行且不受网络拓扑局限等优点。然后,本章提出一种基于无线信道增益的机会调度方案,目标是最大化网络传输速率。最后,本章研究了ANC和TDBC方案中节点的能耗感知功率分配方案,目标是最小化网络传输总功率。(2)第3章提出面向物理层网络编码(Physical-layer Network Coding, PNC)的调度机制,目标是高效地利用无线频谱资源,进而提高网络吞吐量。本章考虑了较双向中继信道传输更复杂的网络传输形式,其中多个节点同时向中继节点发送信息,并基于马尔科夫链进行链路机会调度,其中马氏链的状态为中继节点当前时刻能够接收到源节点发送的数据包个数。接着本章推导了马尔科夫过程各种状态下的转移概率,并对信道状态在对称和非对称情况下所获得的网络吞吐量进行了研究。本章提出的机制在非对称信道状态较对称信道状态在能耗方面更加具有优势,由于大部分网络的信道状态是非对称的,该机制具有广阔的应用场景。(3)尽管传输技术融合可以大幅提高网络性能,但是如何处理不同传输技术应用时相互作用方面的相关研究才刚刚起步。为了最大化网络吞吐量,第4章提出多射频多信道多速率网络编码感知的调度机制,该机制支持不同中继传输方式。本章首先提出一种多速率网络编码感知的调度方案,该方案考虑了网络编码和空分复用的相互作用。由于该问题计算复杂度较高,本章基于列生成算法对该方案进行求解,并提出种启发式算法对列生成的子问题进行简化,目标是进一步降低计算复杂度。接着本章提出一种虚拟链路融合机制,将不采用网络编码但从同一源节点发送到多个口的节点的单播传输扩展为多播传输。最后,将单射频单信道的网络扩展到多射频多信道网络,由于信道分配问题计算复杂度较高,本章采用启发式算法对该问题进行求解。(4)为了对物联网传输中的网络个体进行调度,第5章研究面向社交网络和网络编码的调度机制。本章首先对网络节点的社会属性进行建模,接着提出一种基于经济学双边拍卖模型的中继节点选择策略,该方案鼓励将长距离单跳通信的链路分解为多跳中继转发的链路,降低网络传输干扰的同时将创造出更多链路融合和中继选择机会,从而提高中继节点采用新兴传输方式的可能。随后,本章建立基于社交网络和网络编码的自适应调度传输机制,目标是最大化社会福利和网络吞吐量。由于建立的最优化模型计算复杂度较高,本章最后基于仿生学的萤火虫算法对该问题通过迭代算法进行求解。为了验证和评估上.述机制的性能,本文采用C++和Qualnet仿真软件搭建了仿真平台。仿真结果表明,与现有研究方案相比,本文所提出的方案和算法是有效的,具有更好的网络性能。
[Abstract]:Wireless Mesh networks consist of a group of mobile terminals with dynamic networking capabilities and no infrastructure support. Compared with Ad hoc networks, wireless Mesh networks have attracted much attention from industry and academia in recent years and become the last mile of commercialization due to their advantages of large capacity, high speed, low cost and good scalability. "Wireless broadband hot access technology. With the development of the Internet of Things and the increasing demand for wireless spectrum resources, wireless Mesh network is widely used in metropolitan area network wireless access, vehicle communications and intelligent terminals. With the advent of the era of large data network communications, wireless multimedia services are replacing traditional voice and data communications. Traffic has become the mainstream of network services. How to meet the increasing bandwidth requirements of users is an important issue that researchers need to consider. In addition, the existing research based on wireless Mesh network scheduling mostly assumes that the network channel state information can be known beforehand. However, the real-time changes of wireless network channel environment make it difficult to predict the channel state. Terminals will increasingly reflect people's social attributes, which poses a challenge to communication research in emerging social networks. In view of these problems and challenges, this paper proposes a scheduling mechanism based on channel gain and network coding, a scheduling mechanism based on network coding in Markov model, and a multi-radio frequency multi-channel multi-rate network coding. The main work and innovations of this paper are as follows: (1) Because of the real-time changes of wireless channel environment, channel state is difficult to predict. In chapter 2, wireless channel is proposed based on wireless channel. Gain and Network Coding Opportunity Scheduling and Power Allocation Schemes. Firstly, based on Analog Network Coding (ANC) and Time Division Broadcasting (TDBC) network model, a closed-form expression of outage probability with sliding sampling window is proposed to estimate the gain of wireless channel. The channel gain estimation strategy proposed in this chapter can adapt to the dynamic changes of the network, and has the advantages of simplicity and not limited by the network topology compared with the scheme based on the underlying information collection. Then, this chapter proposes an opportunity scheduling scheme based on the wireless channel gain, aiming to maximize the network transmission rate. The energy-aware power allocation schemes of nodes in ANC and TDBC schemes are studied with the aim of minimizing the total network transmission power. (2) In Chapter 3, a physical-layer network coding (PNC) oriented scheduling mechanism is proposed. The objective is to efficiently utilize wireless spectrum resources and improve network throughput. Relay channel transmission is a more complex form of network transmission, in which multiple nodes send information to the relay node at the same time and schedule links based on Markov chain. The state of Markov chain is the number of packets that the relay node can receive from the source node at the current time. The mechanism proposed in this chapter has more advantages in energy consumption in asymmetric channel state than in symmetric channel state. Because the channel state of most networks is asymmetric, this mechanism has a wide range of applications. (3) Although transmission technology convergence can greatly improve network performance, the research on how to deal with the interaction between different transmission technologies is just beginning. In order to maximize network throughput, Chapter 4 proposes a multi-radio frequency multi-channel multi-rate network coding sensing scheduling mechanism, which supports different relay transmission modes. In this chapter, we first propose a multi-rate network coding-aware scheduling scheme, which considers the interaction between network coding and space division multiplexing. Because of the high computational complexity of this problem, we solve this scheme based on the column generation algorithm, and propose a heuristic algorithm to simplify the sub-problem of column generation. Secondly, a virtual link fusion mechanism is proposed, which extends unicast transmission from the same source node to nodes with multiple ports without network coding to multicast transmission. In this chapter, a heuristic algorithm is used to solve the problem. (4) In order to schedule the individuals in the Internet of Things, Chapter 5 studies the scheduling mechanism for social networks and network coding. The scheme encourages the decomposition of long-distance single-hop communication links into multi-hop relay forwarding links, and reduces network transmission interference while creating more chances for link fusion and relay selection, thus improving the possibility of relay nodes adopting new transmission modes. Subsequently, this chapter establishes adaptive networks and network coding. Scheduling transmission mechanism aims to maximize social welfare and network throughput. Because of the high computational complexity of the optimization model, this chapter finally solves the problem by iteration algorithm based on bionic firefly algorithm. In order to verify and evaluate the performance of the mechanism, this paper uses C++ and Qualnet simulation software to build the simulation. Simulation results show that the proposed scheme and algorithm are effective and have better network performance than existing research schemes.
【学位授予单位】：东北大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN929.5

【相似文献】