报 告 人：赵晓晖 教授 （吉林大学）

报告时间：2016-11-26 AM 9:30-11:30

报告地点：山东大学威海无线传感器网络研究中心（知行南楼404室）

报告人简介：

赵晓晖 博士，吉林大学通信工程学院院长，教授/博士生导师，国务院政府津贴获得者院长。教育部高等学校电子信息类专业教学指导委员会委员。曾任法国巴黎第六大学计算机与通信实验室客座研究员。中国通信学会和电子学会高级会员，IEEE 会员。《通信学报》、《信号处理》、《吉林大学学报信息科学版》和《The Journal of China Universities of Posts and Communications》杂志编委。长期从事认知无线电、信号处理理论及其在通信中的应用领域的科学研究和教学工作。近年来在认知无线电系统的功率控制和自由空间光通信理论等领域取得了若干研究成果，已培养出20多名博士研究生和100多名硕士研究生。近年来在国内外学术刊物和会议上发表SCI/ EI检索学术论文100多篇。已主持完成了国家自然科学基金3项、教育部博士点基金1项、中法、中韩国际合作项目、省部级科研项目多项。

报告内容：

We investigate the robust power allocation issue in orthogonal frequency division multiplexing based cognitive radio networks (CRNs) where the uncertainties are unavoidable. To reduce the performance degradation of system because of the uncertainties, we aim to maximize the data rate of second users (SUs) in considering maximum allowable interference constraints and total power budget of SUs. We design a controller based on a switched affine system with state constraint from a distributed projected dynamic system which provides insight to the behavior of the distributed variational inequality in accordance with the classical distributed convex optimization model for power allocation and dynamic property of the Nash equilibrium. The robust controller design is on the basis of Lyapunov stability theory and linear matrix inequality to better realize the original power allocation from the control perspective. This result is the first of this kind approach for the problem under the control frame, which is more practical for the realization in CRNs. Simulation results are provided to show the validation of the effectiveness of this approach in comparison with iterative water filling algorithm.