首页> 外文会议>International Conference on Ubiquitous and Future Networks >A Machine-Learning-Based Handover Prediction for Anticipatory Techniques in Wi-Fi Networks
【24h】

A Machine-Learning-Based Handover Prediction for Anticipatory Techniques in Wi-Fi Networks

机译:基于机器学习的Wi-Fi网络中预期技术的切换预测

获取原文

摘要

Handover and blind spots in Wi-Fi networks generate temporary interruptions of connection between the devices and the access point, with major quality degradation, for example to video streaming. In this paper we propose a technique to predict the event of handover and blind spots in order to allow the implementation of anticipatory techniques, where connection resources are reallocated or video buffers are filled with low-definition video frames before the connection gets lost. The prediction is based on a machine-learning approach, where the received signal strength indicator (RSSI) is monitored and an upcoming handover is recognized by the pattern of the RSSI over time. Since a number of impairments (different paths followed by the user, different movement speed, fading, noise) affect the RSSI evolution, we resort to a neural-network to learn the peculiarities of each handover and solve the pattern recnonitinn Problem.
机译:Wi-Fi网络中的切换和盲点会导致设备与接入点之间的连接暂时中断,从而严重影响视频流的质量。在本文中,我们提出了一种预测切换和盲点事件的技术,以便实现预期的技术,在该技术中,重新分配连接资源或在连接丢失之前将视频缓冲区填充有低清晰度视频帧。该预测基于机器学习方法,其中监视接收信号强度指示器(RSSI),并且随时间推移RSSI的模式可以识别即将到来的切换。由于许多障碍(用户遵循的不同路径,不同的移动速度,衰落,噪声)会影响RSSI的发展,因此我们求助于神经网络来学习每次切换的特殊性并解决模式识别问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号