Design and Development of RNN Anomaly Detection Model for IoT Networks
Cybersecurity is important today because of the increasing growth of the Internet of Things (IoT), which has resulted in a variety of attacks on computer systems and networks. Cyber security has become an increasingly difficult issue to manage as various IoT devices and services grow. Malicious traf...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9777970/ |
| _version_ | 1811332165951029248 |
|---|---|
| author | Imtiaz Ullah Qusay H. Mahmoud |
| author_facet | Imtiaz Ullah Qusay H. Mahmoud |
| author_sort | Imtiaz Ullah |
| collection | DOAJ |
| description | Cybersecurity is important today because of the increasing growth of the Internet of Things (IoT), which has resulted in a variety of attacks on computer systems and networks. Cyber security has become an increasingly difficult issue to manage as various IoT devices and services grow. Malicious traffic identification using deep learning techniques has emerged as a key component of network intrusion detection systems (IDS). Deep learning methods have been a research focus in network intrusion detection. A Recurrent Neural Network (RNN) is useful in a wide range of applications. First, this paper proposes a novel deep learning model for anomaly detection in IoT networks using a recurrent neural network. Long Short Term Memory (LSTM), BiLSTM, and Gated Recurrent Unit (GRU) techniques are used to implement the proposed model for anomaly detection in IoT networks. A Convolutional Neural Network (CNN) can analyze input features without losing important information, making them particularly well suited for feature learning. Next, a hybrid deep learning model was proposed using convolutional and recurrent neural networks. Finally, a lightweight deep learning model for binary classification was proposed using LSTM, BiLSTM, and GRU based approaches. The proposed deep learning models are validated using NSLKDD, BoT-IoT, IoT-NI, IoT-23, MQTT, MQTTset, and IoT-DS2 datasets. Compared to current deep learning implementations, the proposed multiclass and binary classification model achieved high accuracy, precision, recall, and F1 score. |
| first_indexed | 2024-04-13T16:32:39Z |
| format | Article |
| id | doaj.art-a0f6a20a808149a380303cbfe1ada9f0 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-13T16:32:39Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-a0f6a20a808149a380303cbfe1ada9f02022-12-22T02:39:32ZengIEEEIEEE Access2169-35362022-01-0110627226275010.1109/ACCESS.2022.31763179777970Design and Development of RNN Anomaly Detection Model for IoT NetworksImtiaz Ullah0https://orcid.org/0000-0002-2952-7215Qusay H. Mahmoud1https://orcid.org/0000-0003-0472-5757Department of Electrical, Computer and Software Engineering, Ontario Tech University, Oshawa, CanadaDepartment of Electrical, Computer and Software Engineering, Ontario Tech University, Oshawa, CanadaCybersecurity is important today because of the increasing growth of the Internet of Things (IoT), which has resulted in a variety of attacks on computer systems and networks. Cyber security has become an increasingly difficult issue to manage as various IoT devices and services grow. Malicious traffic identification using deep learning techniques has emerged as a key component of network intrusion detection systems (IDS). Deep learning methods have been a research focus in network intrusion detection. A Recurrent Neural Network (RNN) is useful in a wide range of applications. First, this paper proposes a novel deep learning model for anomaly detection in IoT networks using a recurrent neural network. Long Short Term Memory (LSTM), BiLSTM, and Gated Recurrent Unit (GRU) techniques are used to implement the proposed model for anomaly detection in IoT networks. A Convolutional Neural Network (CNN) can analyze input features without losing important information, making them particularly well suited for feature learning. Next, a hybrid deep learning model was proposed using convolutional and recurrent neural networks. Finally, a lightweight deep learning model for binary classification was proposed using LSTM, BiLSTM, and GRU based approaches. The proposed deep learning models are validated using NSLKDD, BoT-IoT, IoT-NI, IoT-23, MQTT, MQTTset, and IoT-DS2 datasets. Compared to current deep learning implementations, the proposed multiclass and binary classification model achieved high accuracy, precision, recall, and F1 score.https://ieeexplore.ieee.org/document/9777970/Internet of Thingsanomaly detectionrecurrent neural networkconvolutional neural networkLSTMBiLSTM |
| spellingShingle | Imtiaz Ullah Qusay H. Mahmoud Design and Development of RNN Anomaly Detection Model for IoT Networks IEEE Access Internet of Things anomaly detection recurrent neural network convolutional neural network LSTM BiLSTM |
| title | Design and Development of RNN Anomaly Detection Model for IoT Networks |
| title_full | Design and Development of RNN Anomaly Detection Model for IoT Networks |
| title_fullStr | Design and Development of RNN Anomaly Detection Model for IoT Networks |
| title_full_unstemmed | Design and Development of RNN Anomaly Detection Model for IoT Networks |
| title_short | Design and Development of RNN Anomaly Detection Model for IoT Networks |
| title_sort | design and development of rnn anomaly detection model for iot networks |
| topic | Internet of Things anomaly detection recurrent neural network convolutional neural network LSTM BiLSTM |
| url | https://ieeexplore.ieee.org/document/9777970/ |
| work_keys_str_mv | AT imtiazullah designanddevelopmentofrnnanomalydetectionmodelforiotnetworks AT qusayhmahmoud designanddevelopmentofrnnanomalydetectionmodelforiotnetworks |