Research on intelligent routing with VAE-GAN in the internet of body.
The "Internet of Body" is an emerging technology that is centered on the human body and connected to the Internet. It can monitor a variety of human data (such as heart rate, blood oxygen content, etc.) and communicate with digital pills, wearable devices, etc. It has been widely used in t...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2025-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0317698 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850144387043950592 |
|---|---|
| author | Song Qian Tianping Zhang Siping Hu |
| author_facet | Song Qian Tianping Zhang Siping Hu |
| author_sort | Song Qian |
| collection | DOAJ |
| description | The "Internet of Body" is an emerging technology that is centered on the human body and connected to the Internet. It can monitor a variety of human data (such as heart rate, blood oxygen content, etc.) and communicate with digital pills, wearable devices, etc. It has been widely used in the field of medical health. However, when other devices access the Internet of Body on a large scale, there will be load imbalance caused by the difficulty in selecting the optimal route, which will affect the overall throughput and may even fail to transmit and endanger life. The traditional artificial intelligence routing algorithm cannot deal with the low model prediction accuracy and poor generalization ability caused by large noise and small data volume. This paper proposes an artificial intelligence routing algorithm, combines the variational autoencoder (VAE) and the generative adversarial network model (GAN) to construct a VAE-GAN model to generate multiple sets of data to achieve data enhancement on the Internet of Body. The optimization goals are to maximize the throughput of the Internet of Body and minimize the transmission cost. The entire routing problem is expressed as a Markov decision and the optimal transmission path is solved by learning previous historical experience to generate the real-time optimal route. Experiments have shown that this scheme can achieve the optimal route according to the transmission capacity of the real-time path and only requires fewer computing resources. It achieves load balancing of the entire network and avoids network congestion. The average throughput is much higher than that of traditional routing, and the advantage is more obvious under high load. |
| format | Article |
| id | doaj-art-c939a2b08aa74c30ade25e02bdb04d03 |
| institution | OA Journals |
| issn | 1932-6203 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj-art-c939a2b08aa74c30ade25e02bdb04d032025-08-20T02:28:23ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01202e031769810.1371/journal.pone.0317698Research on intelligent routing with VAE-GAN in the internet of body.Song QianTianping ZhangSiping HuThe "Internet of Body" is an emerging technology that is centered on the human body and connected to the Internet. It can monitor a variety of human data (such as heart rate, blood oxygen content, etc.) and communicate with digital pills, wearable devices, etc. It has been widely used in the field of medical health. However, when other devices access the Internet of Body on a large scale, there will be load imbalance caused by the difficulty in selecting the optimal route, which will affect the overall throughput and may even fail to transmit and endanger life. The traditional artificial intelligence routing algorithm cannot deal with the low model prediction accuracy and poor generalization ability caused by large noise and small data volume. This paper proposes an artificial intelligence routing algorithm, combines the variational autoencoder (VAE) and the generative adversarial network model (GAN) to construct a VAE-GAN model to generate multiple sets of data to achieve data enhancement on the Internet of Body. The optimization goals are to maximize the throughput of the Internet of Body and minimize the transmission cost. The entire routing problem is expressed as a Markov decision and the optimal transmission path is solved by learning previous historical experience to generate the real-time optimal route. Experiments have shown that this scheme can achieve the optimal route according to the transmission capacity of the real-time path and only requires fewer computing resources. It achieves load balancing of the entire network and avoids network congestion. The average throughput is much higher than that of traditional routing, and the advantage is more obvious under high load.https://doi.org/10.1371/journal.pone.0317698 |
| spellingShingle | Song Qian Tianping Zhang Siping Hu Research on intelligent routing with VAE-GAN in the internet of body. PLoS ONE |
| title | Research on intelligent routing with VAE-GAN in the internet of body. |
| title_full | Research on intelligent routing with VAE-GAN in the internet of body. |
| title_fullStr | Research on intelligent routing with VAE-GAN in the internet of body. |
| title_full_unstemmed | Research on intelligent routing with VAE-GAN in the internet of body. |
| title_short | Research on intelligent routing with VAE-GAN in the internet of body. |
| title_sort | research on intelligent routing with vae gan in the internet of body |
| url | https://doi.org/10.1371/journal.pone.0317698 |
| work_keys_str_mv | AT songqian researchonintelligentroutingwithvaeganintheinternetofbody AT tianpingzhang researchonintelligentroutingwithvaeganintheinternetofbody AT sipinghu researchonintelligentroutingwithvaeganintheinternetofbody |