为了提高非法入侵行为检测效果，确保船舶物联网安全、平稳运行，提出多模型组合的船舶物联网非法入侵行为检测方法。基于密集连接卷积神经网络、门控循环单元基本原理，构建基于DCCNet-GRU的改进船舶物联网非法入侵检测组合模型，将船舶物联网流量数据作为组合模型的输入，通过一维DCCNet网络获取输入样本数据的空间特征，将其作为GRU的输入，完成其时间维度特征的提取后，再将其输入到全连接层中，实现非法船舶物联网入侵行为类型的识别。实验结果表明，该方法可实现船舶物联网非法入侵行为检测，以Focal为损失函数，增长率为30、网络层数为95时，非法入侵行为检测效果最优。
In order to improve the detection effect of illegal intrusion behavior and ensure the safe and stable operation of the ship Internet of Things, a multi model combination method for detecting illegal intrusion behavior in the ship Internet of Things is proposed. Based on the basic principles of dense connected convolutional neural networks and gated loop units, an improved ship IoT illegal intrusion detection combined model based on DCCNet GRU is constructed. The ship IoT traffic data is used as input to the combined model, and the spatial features of the input sample data are obtained through a one-dimensional DCCNet network as input to GRU. After extracting its temporal features, they are input into the fully connected layer, Identify the types of illegal ship IoT intrusion behaviors. The experimental results show that this method can achieve the detection of illegal intrusion behavior in the ship Internet of Things. With Focal as the loss function, a growth rate of 30, and a network layer of 95, the detection effect of illegal intrusion behavior is optimal.
2023,45(19): 189-192 收稿日期：2023-03-24
DOI：10.3404/j.issn.1672-7649.2023.19.036
分类号：TP391
作者简介：张晓伟(1978-),男,高级工程师,研究方向为计算机科学
参考文献：
[1] 乔楠, 李振兴, 赵国生. XGBoost-RF的物联网入侵检测模型[J]. 小型微型计算机系统, 2022, 43(1): 7-11.
QIAO Nan, LI Zhen-xing, ZHAO Guo-sheng. Intrusion detection model of Internet of Things based on XGBoost-RF[J]. Journal of Chinese Computer Systems, 2022, 43(1): 7-11.
[2] 张娅. 基于深度学习网络的物联网非法入侵识别研究[J]. 微电子学与计算机, 2020, 37(6): 75-78.
ZHANG Ya. Research on Internet of Things illegal intrusion recognition based on deep learning network[J]. Microelectronics & Computer, 2020, 37(6): 75-78.
[3] 胡向东, 周巧. 基于深度学习的工业物联网智能入侵检测[J]. 计算机系统应用, 2020, 29(9): 47-56.
HU Xiang-dong, ZHOU Qiao. IIoT intelligent intrusion detection based on deep learning[J]. Computer Systems & Applications, 2020, 29(9): 47-56.
[4] 周飞菲. 基于支持向量机方法的网络入侵检测实验研究[J]. 西南师范大学学报(自然科学版), 2020, 45(1): 57-61.
ZHOU Fei-fei. Support vector machine method for network intrusion detection[J]. Journal of Southwest China Normal University(Natural Science Edition), 2020, 45(1): 57-61.
[5] 马明艳, 陈伟, 吴礼发. 基于CNN_BiLSTM网络的入侵检测方法[J]. 计算机工程与应用, 2022, 58(10): 116-124.
MA Ming-yan, CHEN Wei, WU Li-fa. Intrusion detection method based on CNN_BiLSTM network[J]. Computer Engineering and Applications, 2022, 58(10): 116-124.
[6] 李俊, 夏松竹, 兰海燕, 等. 基于GRU-RNN的网络入侵检测方法[J]. 哈尔滨工程大学学报, 2021, 42(6): 879-884.
LI Jun, XIA Song-zhu, LAN Hai-yan, et al. Network intrusion detection method based on GRU-RNN[J]. Journal of Harbin Engineering University, 2021, 42(6): 879-884.
