当前位置: X-MOL 学术Int. J. Inf. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Automatic handover execution technique using machine learning algorithm for heterogeneous wireless networks
International Journal of Information Technology Pub Date : 2021-03-01 , DOI: 10.1007/s41870-021-00627-9
Nishatbanu Nayakwadi , Ruksar Fatima

Integrating LTE sub-6 GHz and millimeter wave (mmWave) bands brings great benefit in increasing communication bandwidth, reliability, and better coverage of future smart intelligent network and its applications. However, finding the right mmWave remote radio units (RRUs) is challenging due to coverage blindness of directional beams. Further, the mmWave network depends on edge cloud deployment for satisfying low latency requirement of future smart applications. Along with, reducing energy consumption for handover execution is important due to the battery constraint of IoT (Internet of Things) device. Thus, it is important to reduce the signaling overhead of the handover process. First, this paper presents an efficient handover mechanism between LTE and mmWave; second present automatic handover execution mechanism between LTE and mmWave using a machine learning algorithm. Third presented improved XGBoost classification algorithm for predicting handover success rate using channel information collected through sampling window. Lastly, showed combining machine learning prediction model with standard handover execution model reduces signaling overhead and improves the handover success rate. The experiment is conducted by varying IoT device the result achieved shows XGBoost-based handover execution model achieves much superior performance than existing KNN-based handover execution algorithm.



中文翻译:

使用机器学习算法的异构无线网络自动切换执行技术

整合LTE低于6 GHz频段和毫米波(mmWave)频段,在增加通信带宽,提高可靠性以及更好地覆盖未来智能智能网络及其应用方面带来了巨大的好处。但是,由于定向波束的覆盖盲目性,找到合适的毫米波远程无线电单元(RRU)具有挑战性。此外,mmWave网络依靠边缘云部署来满足未来智能应用程序的低延迟要求。同时,由于物联网(IoT)设备的电池限制,减少用于执行切换的能耗也很重要。因此,重要的是减少切换过程的信令开销。首先,本文介绍了LTE与mmWave之间的高效切换机制;第二种使用机器学习算法的LTE与mmWave之间的自动切换执行机制。第三部分提出了改进的XGBoost分类算法,该算法使用通过采样窗口收集的信道信息来预测切换成功率。最后,展示了将机器学习预测模型与标准切换执行模型相结合可以减少信令开销并提高切换成功率。该实验是通过各种IoT设备进行的,所获得的结果表明,基于XGBoost的移交执行模型比现有的基于KNN的移交执行算法具有更高的性能。展示了将机器学习预测模型与标准移交执行模型相结合可以减少信令开销并提高移交成功率。该实验是通过各种IoT设备进行的,所获得的结果表明,基于XGBoost的移交执行模型比现有的基于KNN的移交执行算法具有更高的性能。展示了将机器学习预测模型与标准移交执行模型相结合可以减少信令开销并提高移交成功率。该实验是通过各种IoT设备进行的,所获得的结果表明,基于XGBoost的移交执行模型比现有的基于KNN的移交执行算法具有更高的性能。

更新日期:2021-03-01
down
wechat
bug