Unsupervised intrusion detection model based on temporal convolutional network
Most existing intrusion detection models rely on long short-term memory (LSTM) networks to consider time-dependencies among data. However, LSTM’s sequential data processing significantly increases computational complexity and memory consumption during training. Therefore, unsupervised intrusion dete...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Beijing Xintong Media Co., Ltd
2025-01-01
|
| Series: | Dianxin kexue |
| Subjects: | |
| Online Access: | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2025001/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823864829202399232 |
|---|---|
| author | LIAO Jinju DING Jiawei FENG Guanghui |
| author_facet | LIAO Jinju DING Jiawei FENG Guanghui |
| author_sort | LIAO Jinju |
| collection | DOAJ |
| description | Most existing intrusion detection models rely on long short-term memory (LSTM) networks to consider time-dependencies among data. However, LSTM’s sequential data processing significantly increases computational complexity and memory consumption during training. Therefore, unsupervised intrusion detection model based on multi-head attention mechanism and temporal convolutional network (UDMT) was proposed. UDMT didn’t rely on LSTM networks. Instead, it used temporal convolutional network and multi-head attention mechanism in the generative adversarial network generator and discriminator networks to enable more computation parallelization, and reduced computational complexity. Moreover, UDMT was capable of detecting both known and zero-day attacks without relying on labeled attack data. In addition, UDMT can adopt different privacy layer modes, and the configuration was flexible to meet the requirements of different detection rates and detection delays. Experiment results show that the proposed UDMT has higher detection rate and lower detection latency than two state-of-the-art intrusion detection models. |
| format | Article |
| id | doaj-art-2545e184b2ef4ae9a6ad08fe788b8b49 |
| institution | Kabale University |
| issn | 1000-0801 |
| language | zho |
| publishDate | 2025-01-01 |
| publisher | Beijing Xintong Media Co., Ltd |
| record_format | Article |
| series | Dianxin kexue |
| spelling | doaj-art-2545e184b2ef4ae9a6ad08fe788b8b492025-02-08T19:00:21ZzhoBeijing Xintong Media Co., LtdDianxin kexue1000-08012025-01-014116417382011722Unsupervised intrusion detection model based on temporal convolutional networkLIAO JinjuDING JiaweiFENG GuanghuiMost existing intrusion detection models rely on long short-term memory (LSTM) networks to consider time-dependencies among data. However, LSTM’s sequential data processing significantly increases computational complexity and memory consumption during training. Therefore, unsupervised intrusion detection model based on multi-head attention mechanism and temporal convolutional network (UDMT) was proposed. UDMT didn’t rely on LSTM networks. Instead, it used temporal convolutional network and multi-head attention mechanism in the generative adversarial network generator and discriminator networks to enable more computation parallelization, and reduced computational complexity. Moreover, UDMT was capable of detecting both known and zero-day attacks without relying on labeled attack data. In addition, UDMT can adopt different privacy layer modes, and the configuration was flexible to meet the requirements of different detection rates and detection delays. Experiment results show that the proposed UDMT has higher detection rate and lower detection latency than two state-of-the-art intrusion detection models.http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2025001/intrusion detection modellong short-term memory networkgenerative adversarial networkmulti-head attention mechanismtemporal convolutional network |
| spellingShingle | LIAO Jinju DING Jiawei FENG Guanghui Unsupervised intrusion detection model based on temporal convolutional network Dianxin kexue intrusion detection model long short-term memory network generative adversarial network multi-head attention mechanism temporal convolutional network |
| title | Unsupervised intrusion detection model based on temporal convolutional network |
| title_full | Unsupervised intrusion detection model based on temporal convolutional network |
| title_fullStr | Unsupervised intrusion detection model based on temporal convolutional network |
| title_full_unstemmed | Unsupervised intrusion detection model based on temporal convolutional network |
| title_short | Unsupervised intrusion detection model based on temporal convolutional network |
| title_sort | unsupervised intrusion detection model based on temporal convolutional network |
| topic | intrusion detection model long short-term memory network generative adversarial network multi-head attention mechanism temporal convolutional network |
| url | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2025001/ |
| work_keys_str_mv | AT liaojinju unsupervisedintrusiondetectionmodelbasedontemporalconvolutionalnetwork AT dingjiawei unsupervisedintrusiondetectionmodelbasedontemporalconvolutionalnetwork AT fengguanghui unsupervisedintrusiondetectionmodelbasedontemporalconvolutionalnetwork |