无线Mesh网络多播联合路由选择与信道分配研究

发布时间：2018-07-26 06:20

【摘要】：无线Mesh网络(Wireless Mesh Networks,WMN)作为一种关键的无线网络接入方式,可以扩展无线接入网的覆盖范围,增加无线接入网的组网灵活性和有效性。由于WMN具有高可靠性、自组织性、自愈性等多种优点,从出现以来一直受到学术界的广泛关注。多接口多信道技术(Multi-Radio Multi-Channel,MRMC)是提升WMN网络性能的一种重要手段,在同一个节点配置多个无线接口,可以增加总体的网络容量。但使用MRMC技术为网络带来了无线干扰增大的问题。WMN中的路由技术通过为数据包选择适当的路径提升网络容量,信道分配技术通过为网络接口分配适当的信道,降低节点之间的干扰,提升网络容量。因此研究WMN中的路由与信道分配技术对解决干扰问题和提升网络性能至关重要。随着无线多媒体等业务的快速发展,WMN中的多播通信需求不断增加。在为WMN进行路由与信道分配时,需要考虑通信的类型,在为使用多播业务的WMN进行路由选择和信道分配时,直接使用用于单播通信的分配方法会导致无法充分发挥多播业务的优势,导致多播吞吐量下降。多播通信时的路由选择通常是构建一个以网关节点为根节点的树形结构,路由选择问题也转化为多播树的构建问题;在为多播通信网络分配信道时,需要充分考虑利用到无线广播优势,提升多播吞吐量。本文研究WMN中多播联合路由选择与信道分配问题。本文为了在得到多播路由与信道分配的最优解,同时降低求解的复杂度,提出一个序列化二进制整数规划(Sequential BIP,SBIP)方法,SBIP将信道分配与路由选择作为两个独立的子问题并且分别进行求解。SBIP为了提高信道频谱的利用率,在使用正交信道的同时,也使用了部分重叠信道,对SBIP的计算复杂度进行了理论分析,结果表明序列化算法可以有效地减小优化过程的计算复杂度,使用NS-3仿真平台验证SBIP的性能,仿真实验表明,SBIP可以得到最优的路由选择结果与信道分配结果,SBIP可以作为其它方法性能的评价标准。SBIP虽然可以在较小的计算复杂度下得到路由选择与信道分配的最优解,但由于其无法在多项式时间内求得结果,无法应用于实际网络。因此,本文提出一种启发式的最小化冲突负载路由选择与信道分配(Min-Collision Load Routing and Channel Assignment,MCLRC)算法。MCLRC可以在多项式时间内得到路由与信道分配结果,可以应用于实际的通信网络。MCLRC的路由选择过程按照离源节点由远及近的顺序为网络决定多播树的转发节点,可以有效地减小接口数的浪费现象;MCLRC的信道分配过程的目的是最小化网络中受干扰影响的带宽,得到的分配结果可以提高网络的总体多播吞吐量。NS-3的仿真结果表明,MCLRC算法性能接近SBIP得到的最优解,且优于MCM与M4算法。
[Abstract]:As a key wireless network access mode, wireless Mesh network (Wireless Mesh network can extend the coverage of wireless access network and increase the flexibility and effectiveness of wireless access network. Because WMN has many advantages such as high reliability, self-organization, self-healing and so on, it has been widely concerned by academic circles since its appearance. Multi-Radio Multi-Channel technique (MRMC) is an important means to improve the performance of WMN networks. Configuring multiple wireless interfaces at the same node can increase the overall network capacity. However, the use of MRMC technology brings the problem of increasing wireless interference. The routing technology in WMN increases the network capacity by selecting the appropriate path for the packet, and the channel allocation technology allocates the appropriate channel for the network interface. Reduce the interference between nodes and improve the network capacity. Therefore, it is very important to study the routing and channel allocation technology in WMN to solve the interference problem and improve the network performance. With the rapid development of wireless multimedia and other services, the demand for multicast communication in WMN is increasing. When routing and channel allocation for WMN, you need to consider the type of communication, when routing and channel allocation for WMN using multicast services, The direct use of the allocation method for unicast communication will lead to the failure to give full play to the advantages of multicast services and lead to the decline of multicast throughput. In multicast communication, routing is usually a tree structure with gateway node as the root node, and the routing problem is also transformed into the construction of multicast tree. It is necessary to take full account of the advantages of wireless broadcasting to improve multicast throughput. In this paper, the problem of multicast joint routing and channel assignment in WMN is studied. In order to obtain the optimal solution of multicast routing and channel assignment, the complexity of the solution is reduced. In this paper, a serialized binary integer programming (Sequential) method is proposed, in which channel assignment and routing are regarded as two independent sub-problems. In order to improve the efficiency of channel spectrum, channel allocation and routing are solved separately. In order to improve the efficiency of the channel spectrum, orthogonal channels are used at the same time. Some overlapping channels are also used to analyze the computational complexity of SBIP. The results show that the serialization algorithm can effectively reduce the computational complexity of the optimization process, and use the NS-3 simulation platform to verify the performance of SBIP. The simulation results show that SBIP can obtain the optimal routing and channel assignment results. SBIP can be used as the evaluation standard for the performance of other methods. SBIP can obtain the optimal solution of routing and channel assignment under lower computational complexity. However, it can not be applied to the real network because it can not get the result in polynomial time. Therefore, this paper proposes a heuristic algorithm for minimizing collision load routing and Channel assignment (MCLRC). MCLRC can obtain routing and channel assignment results in polynomial time. The routing process can be applied to the actual communication network. MCLRC determines the forwarding node of the multicast tree for the network according to the order of distance and proximity of the source node. The purpose of the channel allocation process of MCLRC is to minimize the bandwidth affected by interference in the network. The simulation results show that the performance of MCLRC is close to the optimal solution obtained by SBIP and is superior to that of MCM and M4.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5

【相似文献】