北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2013, Vol. 36 ›› Issue (6): 45-49.doi: 10.13190/j.jbupt.2013.06.010

• 论文 • 上一篇    下一篇

基于神经网络预测CQI的LTE下行调度

吴斌1, 糜正琨1, 王文鼐1, 卢元晨2, 朱泽文1   

  1. 1. 南京邮电大学 宽带无线通信与传感网技术教育部重点实验室, 南京 210003;
    2. 华为软件技术有限公司, 南京 210012
  • 收稿日期:2013-06-19 出版日期:2013-12-31 发布日期:2013-10-08
  • 作者简介:吴摇斌(1974—),男,博士生,E-mail:wubin@njupt.edu.cn;糜正琨(1946—),男,教授,博士生导师.
  • 基金资助:

    国家自然科学基金项目(60872018);国家科技重大专项项目(2011ZX03005-004-03);江苏高校优势学科建设工程资助项目

A LTE Downlink Scheduling Based on CQI Predicted by Neural Network

WU Bin1, MI Zheng-kun1, WANG Wen-nai1, LU Yuan-chen2, ZHU Ze-wen1   

  1. 1. Key Laboratory of Broadband Wireless Communication and Sensor Network Technology (Nanjing University of Posts and Telecommunications), Ministry of Education, Nanjing 210003, China;
    2. Huawei Software Technologies Company Limited, Nanjing 210012, China
  • Received:2013-06-19 Online:2013-12-31 Published:2013-10-08

摘要:

提出一种基于神经网络预测信道质量指示(CQI)的长期演进下行调度方法,设计了基于神经网络的CQI预测模型,使用系统中已获得的CQI历史值来预测调度时刻的当前CQI值,调度器使用此预测值进行调度. 仿真结果表明,该方法对由CQI延迟所导致的系统性能下降有明显改善.

关键词: 3GPP长期演进, 无线资源调度, 信道质量指示预测, 神经网络

Abstract:

Using the available outdated channel quality indicator (CQI), an optimized long term evolution downlink resource allocation method is proposed. The key is a neural network based on CQI prediction model and algorithm, which can predict the current CQI from the outdated CQIs. The scheduler uses the predicted CQI to allocate radio resources and decide on proper modulation and coding scheme. Simulation shows that the proposed prediction model and algorithm have practical effects on solving the degradation of scheduling efficiency due to the outdated CQI.

Key words: 3GPP long term evolution, radio resource scheduling, channel quality indicator prediction, neural network

中图分类号: