Journal of Computer Applications ›› 2022, Vol. 42 ›› Issue (7): 2118-2124.DOI: 10.11772/j.issn.1001-9081.2021040692
• Cyber security • Previous Articles
Ning DONG(), Xiaorong CHENG, Mingquan ZHANG
Received:
2021-04-30
Revised:
2021-08-06
Accepted:
2021-08-10
Online:
2022-07-15
Published:
2022-07-10
Contact:
Ning DONG
About author:
CHENG Xiaorong, born in 1963, Ph. D., professor. Her research interests include network security, big data.Supported by:
通讯作者:
董宁
作者简介:
程晓荣(1963—),女,河北邯郸人,教授,博士,主要研究方向:网络安全、大数据基金资助:
CLC Number:
Ning DONG, Xiaorong CHENG, Mingquan ZHANG. Intrusion detection system with dynamic weight loss function based on internet of things platform[J]. Journal of Computer Applications, 2022, 42(7): 2118-2124.
董宁, 程晓荣, 张铭泉. 基于物联网平台的动态权重损失函数入侵检测系统[J]. 《计算机应用》唯一官方网站, 2022, 42(7): 2118-2124.
Add to citation manager EndNote|Ris|BibTeX
URL: http://www.joca.cn/EN/10.11772/j.issn.1001-9081.2021040692
数据集 | DoS | Probe | U2R | R2L | Normal |
---|---|---|---|---|---|
训练集 | 45 927 | 11 656 | 43 | 995 | 67 351 |
测试集 | 7 459 | 2 421 | 65 | 2 743 | 9 855 |
Tab. 1 NSL-KDD dataset sample distribution
数据集 | DoS | Probe | U2R | R2L | Normal |
---|---|---|---|---|---|
训练集 | 45 927 | 11 656 | 43 | 995 | 67 351 |
测试集 | 7 459 | 2 421 | 65 | 2 743 | 9 855 |
攻击类型 | 数据量 |
---|---|
Normal | 40 073 |
Mirai | 415 677 |
DoS | 59 391 |
ARP | 35 377 |
Scan | 75 266 |
Tab. 2 IoT intrusion detection dataset sample distribution
攻击类型 | 数据量 |
---|---|
Normal | 40 073 |
Mirai | 415 677 |
DoS | 59 391 |
ARP | 35 377 |
Scan | 75 266 |
标签 | 预测 | |||||
---|---|---|---|---|---|---|
DoS | ARP | Mirai | Normal | Scan | 共计 | |
共计 | 17 645 | 937 | 128 661 | 11 184 | 29 573 | 188 000 |
DoS | 17 641 | 0 | 42 | 10 | 0 | 17 693 |
ARP | 0 | 928 | 5 960 | 30 | 3 616 | 10 534 |
Mirai | 0 | 6 | 120 543 | 47 | 4 579 | 125 175 |
Normal | 4 | 0 | 690 | 11 084 | 172 | 11 950 |
Scan | 0 | 3 | 1 426 | 13 | 21 206 | 22 648 |
Tab. 3 Cross-entropy loss function model classification
标签 | 预测 | |||||
---|---|---|---|---|---|---|
DoS | ARP | Mirai | Normal | Scan | 共计 | |
共计 | 17 645 | 937 | 128 661 | 11 184 | 29 573 | 188 000 |
DoS | 17 641 | 0 | 42 | 10 | 0 | 17 693 |
ARP | 0 | 928 | 5 960 | 30 | 3 616 | 10 534 |
Mirai | 0 | 6 | 120 543 | 47 | 4 579 | 125 175 |
Normal | 4 | 0 | 690 | 11 084 | 172 | 11 950 |
Scan | 0 | 3 | 1 426 | 13 | 21 206 | 22 648 |
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DoS | 0.999 7 | 0.999 7 | 0.990 7 | 0.995 1 |
ARP | 0.948 8 | 0.990 3 | 0.088 0 | 0.161 6 |
Mirai | 0.932 1 | 0.936 9 | 0.962 9 | 0.949 7 |
Normal | 0.998 8 | 0.991 0 | 0.927 5 | 0.958 1 |
Scan | 0.947 8 | 0.707 0 | 0.936 3 | 0.805 6 |
Tab. 4 Cross-entropy loss function model evaluation
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DoS | 0.999 7 | 0.999 7 | 0.990 7 | 0.995 1 |
ARP | 0.948 8 | 0.990 3 | 0.088 0 | 0.161 6 |
Mirai | 0.932 1 | 0.936 9 | 0.962 9 | 0.949 7 |
Normal | 0.998 8 | 0.991 0 | 0.927 5 | 0.958 1 |
Scan | 0.947 8 | 0.707 0 | 0.936 3 | 0.805 6 |
标签 | 预测 | |||||
---|---|---|---|---|---|---|
DoS | ARP | Mirai | Normal | Scan | 共计 | |
共计 | 17 715 | 21 532 | 111 380 | 12 168 | 25 205 | 188 000 |
DoS | 17 668 | 8 | 7 | 10 | 0 | 17 693 |
ARP | 30 | 10 187 | 75 | 11 | 231 | 10 534 |
Mirai | 7 | 10 355 | 111 081 | 383 | 3 349 | 125 175 |
Normal | 9 | 115 | 71 | 11 752 | 3 | 11 950 |
Scan | 1 | 867 | 146 | 12 | 21 622 | 22 648 |
Tab. 5 Dynamic weight loss function model classification
标签 | 预测 | |||||
---|---|---|---|---|---|---|
DoS | ARP | Mirai | Normal | Scan | 共计 | |
共计 | 17 715 | 21 532 | 111 380 | 12 168 | 25 205 | 188 000 |
DoS | 17 668 | 8 | 7 | 10 | 0 | 17 693 |
ARP | 30 | 10 187 | 75 | 11 | 231 | 10 534 |
Mirai | 7 | 10 355 | 111 081 | 383 | 3 349 | 125 175 |
Normal | 9 | 115 | 71 | 11 752 | 3 | 11 950 |
Scan | 1 | 867 | 146 | 12 | 21 622 | 22 648 |
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DoS | 0.999 6 | 0.997 3 | 0.998 5 | 0.997 9 |
ARP | 0.937 8 | 0.473 1 | 0.967 0 | 0.635 3 |
Mirai | 0.923 9 | 0.997 3 | 0.887 4 | 0.939 1 |
Normal | 0.996 7 | 0.965 8 | 0.983 4 | 0.974 5 |
Scan | 0.975 4 | 0.857 8 | 0.954 6 | 0.903 6 |
Tab. 6 Dynamic weight loss function model evaluation
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DoS | 0.999 6 | 0.997 3 | 0.998 5 | 0.997 9 |
ARP | 0.937 8 | 0.473 1 | 0.967 0 | 0.635 3 |
Mirai | 0.923 9 | 0.997 3 | 0.887 4 | 0.939 1 |
Normal | 0.996 7 | 0.965 8 | 0.983 4 | 0.974 5 |
Scan | 0.975 4 | 0.857 8 | 0.954 6 | 0.903 6 |
模型 | 时间/μs | 内存/MB | CPU/% | 电量/kWh |
---|---|---|---|---|
CNN | 411.91 | 3.730 | 0.3 | 0.005 |
ANN | 54.09 | 3.144 | 0.3 | 0.004 |
LSTM | 1 355.47 | 1.190 | 0.3 | 0.008 |
CNN-LSTM | 846.10 | 3.870 | 0.7 | 0.007 |
CNN-BiLSTM | 1 066.93 | 3.830 | 0.8 | 0.007 |
Tab. 7 Energy consumption comparison of different intrusion detection systems
模型 | 时间/μs | 内存/MB | CPU/% | 电量/kWh |
---|---|---|---|---|
CNN | 411.91 | 3.730 | 0.3 | 0.005 |
ANN | 54.09 | 3.144 | 0.3 | 0.004 |
LSTM | 1 355.47 | 1.190 | 0.3 | 0.008 |
CNN-LSTM | 846.10 | 3.870 | 0.7 | 0.007 |
CNN-BiLSTM | 1 066.93 | 3.830 | 0.8 | 0.007 |
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DoS | 0.903 7 | 0.961 5 | 0.738 5 | 0.835 3 |
Probe | 0.940 2 | 0.790 2 | 0.603 8 | 0.684 5 |
U2R | 0.997 1 | — | 0 | — |
R2L | 0.878 3 | — | 0 | — |
Normal | 0.747 5 | 0.693 1 | 0.970 4 | 0.770 6 |
Tab. 8 Cross-entropy loss function model evaluation on NSL-KDD dataset
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DoS | 0.903 7 | 0.961 5 | 0.738 5 | 0.835 3 |
Probe | 0.940 2 | 0.790 2 | 0.603 8 | 0.684 5 |
U2R | 0.997 1 | — | 0 | — |
R2L | 0.878 3 | — | 0 | — |
Normal | 0.747 5 | 0.693 1 | 0.970 4 | 0.770 6 |
类型 | 准确率 | 精确率 | 召回率 | F1-measure |
---|---|---|---|---|
DoS | 0.938 9 | 0.966 4 | 0.844 7 | 0.901 4 |
Probe | 0.954 2 | 0.813 1 | 0.742 2 | 0.776 0 |
U2R | 0.995 3 | 0.240 0 | 0.276 9 | 0.257 1 |
R2L | 0.902 4 | 0.944 3 | 0.210 3 | 0.343 9 |
Normal | 0.823 2 | 0.723 6 | 0.963 8 | 0.826 6 |
Tab. 9 Dynamic weight loss function model evaluation on NSL-KDD dataset
类型 | 准确率 | 精确率 | 召回率 | F1-measure |
---|---|---|---|---|
DoS | 0.938 9 | 0.966 4 | 0.844 7 | 0.901 4 |
Probe | 0.954 2 | 0.813 1 | 0.742 2 | 0.776 0 |
U2R | 0.995 3 | 0.240 0 | 0.276 9 | 0.257 1 |
R2L | 0.902 4 | 0.944 3 | 0.210 3 | 0.343 9 |
Normal | 0.823 2 | 0.723 6 | 0.963 8 | 0.826 6 |
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DNN | 0.759 7 | 0.623 9 | 0.509 3 | 0.498 6 |
CNN | 0.754 7 | 0.643 6 | 0.519 2 | 0.514 5 |
DNN FL-NIDS | 0.751 3 | 0.635 1 | 0.487 1 | 0.473 3 |
CNN FL-NIDS | 0.764 3 | 0.765 0 | 0.524 1 | 0.519 6 |
CNN-LSTM | 0.806 9 | 0.836 9 | 0.806 9 | 0.821 6 |
Tab. 10 Classifier comparison
类型 | 准确率 | 精确率 | 召回率 | F1-Measure |
---|---|---|---|---|
DNN | 0.759 7 | 0.623 9 | 0.509 3 | 0.498 6 |
CNN | 0.754 7 | 0.643 6 | 0.519 2 | 0.514 5 |
DNN FL-NIDS | 0.751 3 | 0.635 1 | 0.487 1 | 0.473 3 |
CNN FL-NIDS | 0.764 3 | 0.765 0 | 0.524 1 | 0.519 6 |
CNN-LSTM | 0.806 9 | 0.836 9 | 0.806 9 | 0.821 6 |
1 | AL-GARADI M A, MOHAMED A, AL-ALI A K, et al. A survey of machine and deep learning methods for Internet of Things (IoT) security[J]. IEEE Communications Surveys and Tutorials, 2020, 22(3):1646-1685. 10.1109/comst.2020.2988293 |
2 | CASSALES G W, SENGER H, DE FARIA E R, et al. IDSA-IoT: an intrusion detection system architecture for IoT networks[C]// Proceedings of the 2019 IEEE Symposium on Computers and Communications. Piscataway: IEEE, 2019: 1-7. 10.1109/iscc47284.2019.8969609 |
3 | ULLAH I, MAHMOUD Q H. A scheme for generating a dataset for anomalous activity detection in IoT networks[C]// Proceedings of the 2020 Canadian Conference on Artificial Intelligence, LNCS 12109. Cham: Springer, 2020: 508-520. 10.3390/electronics9030530 |
4 | THAMILARASU G, CHAWLA S. Towards deep-learning-driven intrusion detection for the Internet of Things[J]. Sensors, 2019, 19(9): No.1977. 10.3390/s19091977 |
5 | WANG Z D, LIU Y D, HE D J, et al. Intrusion detection methods based on integrated deep learning model[J]. Computers and Security, 2021, 103: No.102177. 10.1016/j.cose.2021.102177 |
6 | SHONE N, NGOC T N, PHAI V D, et al. A deep learning approach to network intrusion detection[J]. IEEE Transactions on Emerging Topics in computational Intelligence, 2018, 2(1): 41-50. 10.1109/tetci.2017.2772792 |
7 | HASSAN M M, GUMAEI A, ALSANAD A, et al. A hybrid deep learning model for efficient intrusion detection in big data environment[J]. Information Sciences, 2020, 513: 386-396. 10.1016/j.ins.2019.10.069 |
8 | ALMIANI M, AbuGHAZLEH A, AL-RAHAYFEH A, et al. Deep recurrent neural network for IoT intrusion detection system[J]. Simulation Modelling Practice and Theory, 2020, 101: No.102031. 10.1016/j.simpat.2019.102031 |
9 | VASAN D, ALAZAB M, VENKATRAMAN S, et al. MTHAEL: cross-architecture IoT malware detection based on neural network advanced ensemble learning[J]. IEEE Transactions on Computers, 2020, 69(11): 1654-1667. 10.1109/tc.2020.3015584 |
10 | LI Y M, XU Y Y, LIU Z, et al. Robust detection for network intrusion of industrial IoT based on multi-CNN fusion[J]. Measurement, 2020, 154: No.107450. 10.1016/j.measurement.2019.107450 |
11 | 刘辉,张俊鹏,李清荣. 多尺度卷积与动态权重代价函数的全卷积网络工业烟尘目标分割[J]. 计算机辅助设计与图形学学报, 2020, 32(12):1898-1909. 10.3724/SP.J.1089.2020.18215 |
LIU H, ZHANG J P, LI Q R. Industrial smoke target segmentation based on fully convolutional networks with multiscale convolution and dynamic weight loss function[J]. Journal of Computer-Aided Design and Computer Graphics, 2020, 32(12): 1898-1909. 10.3724/SP.J.1089.2020.18215 | |
12 | 唐小棠. 基于机器学习的入侵检测及其在物联网安全的应用[D]. 上海:上海交通大学, 2019:70-79. |
TANG X T. Learning-based intrusion detection methods and their application to IoT security[D]. Shanghai: Shanghai Jiao Tong University, 2019:70-79. | |
13 | 李超,柴玉梅,南晓斐,等. 基于深度学习的问题分类方法研究[J]. 计算机科学, 2016, 43(12):115-119. 10.11896/j.issn.1002-137X.2016.12.020 |
LI C, CHAI Y M, NAN X F, et al. Research on problem classification method based on deep learning[J]. Computer Science, 2016, 43(12): 115-119. 10.11896/j.issn.1002-137X.2016.12.020 | |
14 | CHAWLA N V, BOWYER K W, HALL L O, et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16: 321-357. 10.1613/jair.953 |
15 | 潘建国,李豪. 基于实用拜占庭容错的物联网入侵检测方法[J]. 计算机应用, 2019, 39(6):1742-1746. 10.11772/j.issn.1001-9081.2018102096 |
PAN J G, LI H. Intrusion detection approach for IoT based on practical Byzantine fault tolerance[J]. Journal of Computer Applications, 2019, 39(6): 1742-1746. 10.11772/j.issn.1001-9081.2018102096 | |
16 | MULYANTO M, FAISAL M, PRAKOSA S W, et al. Effectiveness of focal loss for minority classification in network intrusion detection systems[J]. Symmetry, 2021, 13(1): No.4. 10.3390/sym13010004 |
[1] | Yaru HAN, Lianshan YAN, Tao YAO. Deep hashing retrieval algorithm based on meta-learning [J]. Journal of Computer Applications, 2022, 42(7): 2015-2021. |
[2] | Wanjun LIU, Jiaming WANG, Haicheng QU, Libing DONG, Xinyu CAO. Music genre classification algorithm based on attention spectral-spatial feature [J]. Journal of Computer Applications, 2022, 42(7): 2072-2077. |
[3] | Zhenyu WANG, Lei ZHANG, Wenbin GAO, Weiming QUAN. Human activity recognition based on progressive neural architecture search [J]. Journal of Computer Applications, 2022, 42(7): 2058-2064. |
[4] | Jie ZHANG, Shanshan XU, Lingyun YUAN. Internet of things access control model based on blockchain and edge computing [J]. Journal of Computer Applications, 2022, 42(7): 2104-2111. |
[5] | Yang ZHANG, Jiangbo HAO. Malicious code detection method based on attention mechanism and residual network [J]. Journal of Computer Applications, 2022, 42(6): 1708-1715. |
[6] | Shan SU, Yang ZHANG, Dongwen ZHANG. Coupling related code smell detection method based on deep learning [J]. Journal of Computer Applications, 2022, 42(6): 1702-1707. |
[7] | Yumin HAN, Xiaoyan HAO. Material entity recognition based on subword embedding and relative attention [J]. Journal of Computer Applications, 2022, 42(6): 1862-1868. |
[8] | Lei YANG, Hongdong ZHAO, Kuaikuai YU. End-to-end speech emotion recognition based on multi-head attention [J]. Journal of Computer Applications, 2022, 42(6): 1869-1875. |
[9] | Guangkai LIAO, Zheng ZHANG, Zhiguo SONG. Convolutional network-based vehicle re-identification combining wavelet features and attention mechanism [J]. Journal of Computer Applications, 2022, 42(6): 1876-1883. |
[10] | Meng YU, Wentao HE, Xuchuan ZHOU, Mengtian CUI, Keqi WU, Wenjie ZHOU. Review of recommendation system [J]. Journal of Computer Applications, 2022, 42(6): 1898-1913. |
[11] | Jia LI, Yuanlin ZHENG, Kaiyang LIAO, Haojie LOU, Shiyu LI, Zehao CHEN. No-reference image quality assessment algorithm based on saliency deep features [J]. Journal of Computer Applications, 2022, 42(6): 1957-1964. |
[12] | Zhipei YANG, Sheng DING, Li ZHANG, Xinyu ZHANG. Anchor-free remote sensing image detection method for dense objects with rotation [J]. Journal of Computer Applications, 2022, 42(6): 1965-1971. |
[13] | Xiaoyong BIAN, Xiongjun FEI, Chunfang CHEN, Dongdong KAN, Sheng DING. Joint 1-2-order pooling network learning for remote sensing scene classification [J]. Journal of Computer Applications, 2022, 42(6): 1972-1978. |
[14] | Jing JIANG, Yu CHEN, Jieping SUN, Shenggen JU. Integrating posterior probability calibration training into text classification algorithm [J]. Journal of Computer Applications, 2022, 42(6): 1789-1795. |
[15] | Min WEN, Rongcun WANG, Shujuan JIANG. Source code vulnerability detection based on relational graph convolution network [J]. Journal of Computer Applications, 2022, 42(6): 1814-1821. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||