• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Computer Vision, Image and Deep Learning >Application of Convolutional Neural Network in multi-channel Scenario D2D Communication Transmitting Power Control
【24h】

Application of Convolutional Neural Network in multi-channel Scenario D2D Communication Transmitting Power Control

机译：卷积神经网络在多通道场景D2D通信传输功率控制中的应用

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In order to optimize the communication performance of the underlying D2D communication in a multichannel scenario by adjusting the user's transmit power, we present a power control scheme of D2D communication in multichannel scenario based on Convolutional Neural Network (CNN). By using the transmit power of CNN self-learning D2D transmitter, we can maximize the spectrum efficiency of D2D receiver, which is helpful to consider the channel state of each link. In the case of optimizing transmitting power control, conventional methods need to run multiple iterating process to find out the solution of a complex optimizing problem. But deep-learning methods, which is driven by data, can find out a solution with extremely low time consumption by well-trained artificial neural network. In this paper, we note that the input of the neural network is channel status of users and the output is the strategy of transmitting power control. Simulations show that CNN can find out the transmitting power control plan which is closed to the conventional method with low time consumption.
机译：为了通过调整用户的发射功率来优化多声道场景中的基础D2D通信的通信性能，我们在基于卷积神经网络（CNN）的多通道场景中的D2D通信的功率控制方案。通过使用CNN自学习D2D发射器的发射功率，我们可以最大化D2D接收器的频谱效率，这有助于考虑每个链路的信道状态。在优化发射功率控制的情况下，传统方法需要运行多个迭代过程以找出复杂优化问题的解决方案。但是，由数据驱动的深度学习方法，可以通过训练有素的人工神经网络找到具有极低时间消耗的解决方案。在本文中，我们注意到神经网络的输入是用户的信道状态，输出是传输功率控制的策略。模拟表明，CNN可以找出与具有低时间消耗的传统方法关闭的发射功率控制计划。

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