基于贝叶斯网络的认知引擎设计与重配置

摘要

以往的通信行为指导系统未来通信，以满足用户需求并适应环境变化，是认知无线电系统的核心所在，为此提出了一种基于贝叶斯网络的认知引擎，用于解决在复杂多变的电磁环境与用户需求条件下，认知无线电系统参数自适应调整的问题。通过对系统过去通信行为样本数据，进行结构学习和参数学习建立认知引擎，将系统当前环境状态和用户需求信息经预处理作为推理的证据，应用引擎决策出系统此时最佳的工作参数，完成系统参数重构。本文利用OPNET工具建立一个移动无线网络完成仿真实验，仿真结果表明该认知引擎能有效地使移动无线网络适应环境变化，改善端到端通信性能，进一步验证了建模方法的可行性。%The past communication behaviors that guide the system communication in the future to satisfy the requirements of users and adapt to the changes of environment are the core part of cognitive radio system. In this paper, a cognitive engine based on Bayesian network is proposed to solve the parameters self-adaptive-adjusting problem of cognitive radio system under the complicated and highly varying radio environment and user requirement. Through structure learning and parameter learning of the sample data from the past communication behaviors, cognitive engine is established. The states of radio environment and requirements of users are made as inference evidences by data preprocessing, and the cognitive engine is used to make decision of the configuration parameters of com-munication system, and then the reconfiguration system is completed. A mobile wireless network is modeled to finish reconfiguration simulation using OPNET tool in this paper. Simulation results show that the proposed cognitive engine can make the wireless mobile network adapt to environment and effectively improve end-to-end communication performance. The feasibility of the method to model cognitive engine with Bayesian network is validated in this paper.