| Title |
Abnormal human behavior detection based on VAE-LSTM hybrid model in WiFi CSI with PCA |
| ID_Doc |
44380 |
| Authors |
Kim, Y; Kim, SC |
| Title |
Abnormal human behavior detection based on VAE-LSTM hybrid model in WiFi CSI with PCA |
| Year |
2023 |
| Published |
|
| DOI |
10.1109/ICOIN56518.2023.10048984 |
| Abstract |
Recently, It is easy to find network access points(APs), which can be used for more than simply connecting devices to the Internet. For example, the waveform of a WiFi signal changes when a human action is performed between the two APs. In previous research, we demonstrated how changes in an electric wave affect the channel state information of a signal and how deep learning can utilize this information to detect and predict human behavior. In this paper, we proposed a method to detect human behavior. The proposed method improves the performance of detection of human behavior and effective in a changing environment. We found that using a VAE-LSTM hybrid model with PCA is useful in terms of detecting abnormal human behavior Experimental results demonstrate that the proposed method can detect general abnormal behavior with >-79% overall precision in a changing environment. |
| Author Keywords |
LSTM; VAE; CNN; CSI; autoencoder; PCA; RNN; Smart City; IOT |
| Index Keywords |
Index Keywords |
| Document Type |
Other |
| Open Access |
Open Access |
| Source |
Conference Proceedings Citation Index - Science (CPCI-S) |
| EID |
WOS:000981938900150 |
| WoS Category |
Computer Science, Information Systems; Computer Science, Theory & Methods; Engineering, Electrical & Electronic; Telecommunications |
| Research Area |
Computer Science; Engineering; Telecommunications |
| PDF |
|