ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion Detection

Network intrusion detection systems can identify intrusion behavior in a network by analyzing network traffic data. It is challenging to detect a very small proportion of intrusion data from massive network traffic and identify the attack class in intrusion detection tasks. Many existing intrusion d...

Full description

Saved in:
Bibliographic Details
Main Authors: Bin Li, Jie Li, Mingyu Jia
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/5/1382
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850222691144957952
author Bin Li
Jie Li
Mingyu Jia
author_facet Bin Li
Jie Li
Mingyu Jia
author_sort Bin Li
collection DOAJ
description Network intrusion detection systems can identify intrusion behavior in a network by analyzing network traffic data. It is challenging to detect a very small proportion of intrusion data from massive network traffic and identify the attack class in intrusion detection tasks. Many existing intrusion detection studies often fail to fully extract the spatial features of network traffic and make reasonable use of temporal features. In this paper, we propose ADFCNN-BiLSTM, a novel deep neural network for network intrusion detection. ADFCNN-BiLSTM uses deformable convolution and an attention mechanism to adaptively extract the spatial features of network traffic data, and it pays attention to the important features from both channel and spatial perspectives. It uses BiLSTM to mine the temporal features from the traffic data and employs the multi-head attention mechanism to allow the network to focus on the time-series information related to suspicious traffic. In addition, ADFCNN-BiLSTM addresses the issue of class imbalance during the training process at both the data level and algorithm level. We evaluated the proposed ADFCNN-BiLSTM on three standard datasets, i.e., NSL-KDD, UNSW-NB15, and CICDDoS2019. The experimental results show that ADFCNN-BiLSTM outperforms the state-of-the-art model in terms of accuracy, detection rate, and false-positive rate.
format Article
id doaj-art-47f93ef6932c417a9b6d9f593c5385bd
institution OA Journals
issn 1424-8220
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-47f93ef6932c417a9b6d9f593c5385bd2025-08-20T02:06:15ZengMDPI AGSensors1424-82202025-02-01255138210.3390/s25051382ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion DetectionBin Li0Jie Li1Mingyu Jia2School of Computer Science, Northeast Electric Power University, Jilin 132012, ChinaSchool of Computer Science, Northeast Electric Power University, Jilin 132012, ChinaSchool of Computer Science, Northeast Electric Power University, Jilin 132012, ChinaNetwork intrusion detection systems can identify intrusion behavior in a network by analyzing network traffic data. It is challenging to detect a very small proportion of intrusion data from massive network traffic and identify the attack class in intrusion detection tasks. Many existing intrusion detection studies often fail to fully extract the spatial features of network traffic and make reasonable use of temporal features. In this paper, we propose ADFCNN-BiLSTM, a novel deep neural network for network intrusion detection. ADFCNN-BiLSTM uses deformable convolution and an attention mechanism to adaptively extract the spatial features of network traffic data, and it pays attention to the important features from both channel and spatial perspectives. It uses BiLSTM to mine the temporal features from the traffic data and employs the multi-head attention mechanism to allow the network to focus on the time-series information related to suspicious traffic. In addition, ADFCNN-BiLSTM addresses the issue of class imbalance during the training process at both the data level and algorithm level. We evaluated the proposed ADFCNN-BiLSTM on three standard datasets, i.e., NSL-KDD, UNSW-NB15, and CICDDoS2019. The experimental results show that ADFCNN-BiLSTM outperforms the state-of-the-art model in terms of accuracy, detection rate, and false-positive rate.https://www.mdpi.com/1424-8220/25/5/1382network intrusion detectionbidirectional long short-term memorydeformable convolutionattention mechanism
spellingShingle Bin Li
Jie Li
Mingyu Jia
ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion Detection
Sensors
network intrusion detection
bidirectional long short-term memory
deformable convolution
attention mechanism
title ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion Detection
title_full ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion Detection
title_fullStr ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion Detection
title_full_unstemmed ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion Detection
title_short ADFCNN-BiLSTM: A Deep Neural Network Based on Attention and Deformable Convolution for Network Intrusion Detection
title_sort adfcnn bilstm a deep neural network based on attention and deformable convolution for network intrusion detection
topic network intrusion detection
bidirectional long short-term memory
deformable convolution
attention mechanism
url https://www.mdpi.com/1424-8220/25/5/1382
work_keys_str_mv AT binli adfcnnbilstmadeepneuralnetworkbasedonattentionanddeformableconvolutionfornetworkintrusiondetection
AT jieli adfcnnbilstmadeepneuralnetworkbasedonattentionanddeformableconvolutionfornetworkintrusiondetection
AT mingyujia adfcnnbilstmadeepneuralnetworkbasedonattentionanddeformableconvolutionfornetworkintrusiondetection