Authors: Liang Tang; Zhichao Wang; Jun Huang; Li Bian
Addresses: Department of Information Engineering, Suqian College, Suqian, 223800, China ' Department of Information Engineering, Suqian College, Suqian, 223800, China ' Department of Information Engineering, Suqian College, Suqian, 223800, China; Wireless Mobi Solution, 9920 Pacific Heights Blvd, San Diego, CA 92121, USA ' Department of Information Engineering, Suqian College, Suqian, 223800, China
Abstract: When determining the location of candidate wireless mesh routers, it is easy to ensure whether each mesh client node is within the communication range of candidate routers, but it is difficult to ensure that selected mesh routers from the candidate routers can constitute a connected network. In addition, the locations of the candidate router can also be placed too much, but too many candidate routers will affect the performance of network. According to the existing problem, we propose that all user terminals must be within at least one wireless router's coverage, and the least mesh routers to cover all of the user's terminals. Therefore, the selected mesh routers connected with each other, but usually the candidate routers only form several local connected sub-nets without connecting with each other. So making the local connective network communicating with each other is a subsequent work. Using the method of combining the genetic algorithm and the minimum spanning tree algorithm, the problem that candidate mesh routers can construct a general connected network was solved and the least selected wireless Mesh routers connected with each other and all mesh clients are covered. The feasibility and practicability of the proposed method are verified by MATLAB simulation.
Keywords: wireless mesh network; WMN; mesh router deployment; genetic algorithm; connected component; minimum spanning tree.
International Journal of Internet Protocol Technology, 2019 Vol.12 No.1, pp.44 - 50
Received: 17 May 2018
Accepted: 28 Jul 2018
Published online: 25 Mar 2019 *