Channel Assignment Algorithm Based On Discrete BFO For Multi-Channel Wireless Monitoring Networks

In wireless networks (WLAN,WMN,Wimax,Adhoc etc), wireless sniffers are distributed in a region to collect the communication information of users in real time. It can be used for fault diagnosis and resource management. These are important implications for protecting the network security, improving the network performance and enhancing the user experience. As the number of the sniffers is limited, it’s necessary to optimize the schedule of each sniffer’s hardware resources (e.g. communication channel) to maximize the monitoring coverage of users and the information collected, so as to maximize the network’s Quality of Monitoring (QoM). This paper researches the optimization of the sniffer’s channel scheduling, and put forward a channel assignment algorithm based on Discrete Bacterial Foraging Optimization(DBFO) for multi-radio multi-channel wireless monitoring network. Through theoretical derivation and simulation results prove the effectiveness of the method. The main problem studied in this paper and its major innovations are summarized as follows.Firstly, this paper focuses on the multi-channel assignment problem of multi-radio sniffers and establish a corresponding mathematical model. We extend the previous single-radio sniffers’ channel assignment to the multi-radio field. Based on the multi-radio multi-channel wireless monitoring features, we used the indirect bipartite graph to describe the multi-radio wireless monitoring issues. We map the assignment relationship to a two dimensional coding table, effectively transform the channel assignment into coding problem. This paper describes the monitoring quality of both single sniffer and the entire wireless monitoring network in formulation. And the QoM is defined as the performance evaluation index of channel assignment.Secondly, this paper proposes a solution based on DBFO to solve the channel assignment problem for multi-radio sniffers. The traditional continuous BFO algorithm has the characteristics of fine search and global optimization. Based on that, we discrete the dimensional position into a binary 0 or 1, which convert this problem into an encoding one and finally get the optimal solution. This paper analyzes and proves that the DBFO algorithm can convergent to the optimal solution with a probability of 1. We also consider about probability of the dimensional position update. Both simulation and real tests prove the effectiveness of the DBFO algorithm.The research in this paper extends the existing single radio channel assignment problem into more practical multi-radio one. The theory and method proposed in this paper has a role in promoting and reference value for wireless monitoring network technology.