首页 | 官方网站   微博 | 高级检索  
     

基于免疫神经网络的陀螺仪漂移预测
引用本文:蔡 曦,胡昌华,刘炳杰. 基于免疫神经网络的陀螺仪漂移预测[J]. 计算机工程, 2007, 33(24): 237-238
作者姓名:蔡 曦  胡昌华  刘炳杰
作者单位:[1]中国科学院计算技术研究所,北京100080 [2]诺基亚中国研究中心,北京100013
基金项目:国家自然科学基金 , 诺基亚中国研究中心基金
摘    要:BP神经网络可用于预测陀螺飘移误差,但容易陷入局部极值,训练速度很慢。针对上述缺点,该文提出了一种基于免疫算法的神经网络,以样本输出为抗原、神经网络权值矩阵为抗体,通过克隆、变异、抑制等步骤找到最优抗体,将最优抗体用于陀螺仪漂移预测。仿真试验显示,免疫训练算法能有效优化网络权值,基于该模型的漂移预测精度较高。

关 键 词:免疫算法  神经网络  BP算法  陀螺仪漂移  预测
文章编号:1000-3428(2007)24-0004-03
修稿时间:2006-12-20

Gyroscopic Drift Forecasting Based on Immune Neural Network
JIANG Jian,LIU Min,XU Gang,MA Jian,ZHANG Dong-mei. Gyroscopic Drift Forecasting Based on Immune Neural Network[J]. Computer Engineering, 2007, 33(24): 237-238
Authors:JIANG Jian  LIU Min  XU Gang  MA Jian  ZHANG Dong-mei
Affiliation:1. Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100080; 2. Nokia Research Center(China), Beijing 100013
Abstract:This paper introduces a novel retransmission scheme for hybrid automatic repeat request(HARQ) based on priority,this method is used for asynchronous HARQ of multi-user N channel stop and wait mechanism.Because video transmission over wireless is sensitive to packet delay,in order to reduce packet delay retransmission,it calculates the priority based on retransmission waiting time and regulation the order about transmission,transmits packet which has longer waiting time with higher priority.According to theoretical analysis and simulation about this scheme,it can effectively reduce the end to end delay and pack loss for video retransmission over wireless channel under the circumstance of bad channel status so as to meet QoS assurances.
Keywords:hybrid automatic repeat request(HARQ)  priority  multi-user N channel stop and wait
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《计算机工程》浏览原始摘要信息
点击此处可从《计算机工程》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号