基于深度学习的无线频谱感知算法优化
Optimization of Wireless Spectrum Sensing Algorithm Based on Deep Learning
王玉真1
作者信息
- 1. 聊城市技师学院,山东聊城 252000
- 折叠
摘要
文章提出一种基于深度学习的无线频谱感知方法,并引入L1 正则化方法进行优化.首先,分析无线频谱感知的基本原理;其次,设计基于循环神经网络(Recurrent Neural Network,RNN)的频谱感知模型,并对传统RNN模型进行L1 正则化优化;最后,利用单输入多输出(Single Input Multiple Output,SIMO)协同频谱感知场景数据集进行实验验证.实验结果表明,相较于传统RNN方法,优化后的RNN模型在准确率、召回率以及F1 值等指标上均实现了更好的性能表现,证明了文章提出的方法在提高频谱感知性能方面具有显著的优势.
Abstract
This paper proposes a wireless spectrum sensing method based on deep learning,and introduces L1 regularization method for optimization.Firstly,the basic principle of wireless spectrum sensing is analyzed.Secondly,a spectrum sensing model based on Recurrent Neural Network(RNN)is designed,and the traditional RNN model is optimized by L1 regularization.Finally,the experimental verification is carried out by using Single Input Multiple Output(SIMO)collaborative spectrum sensing scene data set.The experimental results show that compared with the traditional RNN method,the optimized RNN model achieves better performance in terms of accuracy,recall and F1 value,which proves that the proposed method has significant advantages in improving spectrum sensing performance.
关键词
深度学习/频谱感知/长短期记忆模型/L1正则化Key words
deep learning/spectrum sensing/long short-term memory model/L1 regularization引用本文复制引用
出版年
2024
NETL
NSTL
万方数据