Safe positioning based on CNN and LSTM for 5G wireless networks

Lu Chen; Guan Mingxiang; Zhou Jianming; Wu Naixing; Gan Yuxi; Tang Hui

doi:10.1117/12.2657201

28 November 2022 Safe positioning based on CNN and LSTM for 5G wireless networks

Lu Chen, Guan Mingxiang, Zhou Jianming, Wu Naixing, Gan Yuxi, Tang Hui

Proceedings Volume 12503, International Conference on Network Communication and Information Security (ICNCIS 2022); 125030J (2022) https://doi.org/10.1117/12.2657201
Event: International Conference on Network Communication and Information Security (ICNCIS 2022), 2022, Qingdao, China

Abstract

This paper presents a robust 5G wireless networks visual safety positioning model, which combines CNN (Convolutional Neural Network) and LSTM (Long Short-Term Memory) networks and can solve the sequence problem. Two parallel full connection layers are added to the output layer of the network to regress the RGB images to obtain the 3D position and 3D direction of the 5G wireless networks. Because the data set of each scene is small, the method of transfer learning is used in the training. The model has the best positioning result in the chess scene on the 7senses dataset, with the positioning error of 0.21m and 7.52°, and the positioning error in the seven scenes is 0.31m and 10.35°. This method can achieve good positioning effect in indoor positioning.

Citation Download Citation

Lu Chen, Guan Mingxiang, Zhou Jianming, Wu Naixing, Gan Yuxi, and Tang Hui "Safe positioning based on CNN and LSTM for 5G wireless networks", Proc. SPIE 12503, International Conference on Network Communication and Information Security (ICNCIS 2022), 125030J (28 November 2022); https://doi.org/10.1117/12.2657201

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available