无线电工程2026，Vol.56Issue(3)：436-443,8.DOI:10.3969/j.issn.1003-3106.2026.03.006
基于轻量级卷积神经网络的5G随机接入信道检测增强方法
Enhanced 5G Random Access Channel Detection Method Based on Lightweight Convolutional Neural Network
摘要
Abstract
The Physical Random Access Channel(PRACH)plays an important role in the 5G system,and PRACH detection is a key issue in the field of wireless signal processing.In a low signal to noise ratio environment,the traditional PRACH detection algorithm is prone to noise interference,resulting in a reduced detection rate.To improve the detection performance,an enhanced PRACH detection method based on deep learning is proposed.Based on the classical correlation algorithm,this method introduces a lightweight Convolutional Neural Network(CNN)constructed by depthwise separable convolution to replace the traditional fixed threshold decision mechanism.Experimental verification shows that,under the condition of maintaining the same false alarm rate,the correct detection rate of this method is superior to that of the traditional threshold decision method.关键词
物理随机接入信道/深度学习/轻量级模型/用户检测Key words
PRACH/deep learning/lightweight model/user detection分类
信息技术与安全科学引用本文复制引用
孔欢,刘奕彤,李卓航,孙宇楠,杨鸿文..基于轻量级卷积神经网络的5G随机接入信道检测增强方法[J].无线电工程,2026,56(3):436-443,8.基金项目
应急通信装备创新揭榜挂帅(E08-01) Emergency Communication Equipment Innovation Uncover-the-rank and Assume-the-role(E08-01) （E08-01）
