适用于认知无线传感器网络的高效频谱分配方法

广泛工作在ISM(Industrial,Scientific and Medical)频段的无线传感器网络面临严重的频谱稀缺问题。在无线通信中,动态频谱分配被认为是提高频谱效能的重要途径。针对典型集中式管理的认知无线传感器网络设计了基于图着色结合负载强度的高效频谱分配算法。首先在协议干扰模型下确保频谱资源在空间上充分利用,随后综合考虑节点负载强度与公平性建立优化模型并按照乘子法加拟牛顿法框架求解最优值。仿真实验表明算法与固定频谱分配方式和传统自适应频谱分配算法相比,在系统吞吐量和节点缓冲区队列长度两个关键性能指标上具有显著优势。

An Efficient Spectrum Allocation Method for Cognitive Wireless Sensor Network

WSN widely working in ISM spectrum band is facing serious problem of spectrum scarcity,in wireless communication,Dynamic Spectrum Allocation(DSA)is considered as an important way to improve spectrum utility. In this paper,an efficient spectrum allocation algorithm based on graph coloring combining with users’load intensity for the typical centralized cognitive WSN is designed. To start with,we ensure the frequency resource fully reused among users on the geographical space. Then load intensity and fairness of users are integrated together to set a optimization model,it is solved through multiplier plus quasi-Newton algorithm structure and obtain and optimal value. Simulation experiments demonstrate that our algorithm has the remarked superiority in the system throughput and node buffer queue length (the two major optimizing objectives)when compare with the fixed spectrum allocation method and the traditional adaptive spectrum allocation algorithm.