A Network Intrusion Detection Method Based on Stacked Autoencoder and LSTM

Yan, Yu; Qi, Lin; Wang, Jie; Lin, Yun; Lei, Chen

doi:10.1109/icc40277.2020.9149384

Cited by 48 publications

(10 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A single LSTM unit is much more complex than a traditional neural unit. It has four gates: input gate, output gate, forget gate, and cell gate [35]. The LSTM cell takes input feature

x_{t}

along with cell state

c_{t - 1}

and hidden state

h_{t - 1}

from previous LSTM units and outputs the current cell state

c_{t}

and hidden state

h_{t}

.…”

Section: Proposed Ids Methodologymentioning

confidence: 99%

Exploring cybersecurity issues in 5G enabled electric vehicle charging station with deep learning

Basnet

Ali

2021

IET Generation Trans & Dist

View full text Add to dashboard Cite

The surging usage of electric vehicles (EVs) demand the robust deployment of trustworthy electric vehicle charging station (EVCS) with millisecond range latency and massive machine to machine communications where 5G could act. However, 5G suffers from inherent protocols, hardware, and software vulnerabilities that seriously threaten the communicating entities' cyber-physical security. To overcome these limitations in the EVCS system, this paper analyses the impact of False Data Injection (FDI) and Distributed Denial of Services (DDoS) attacks on the operation of EVCS. This work is an extension of the previously published conference paper about the EVCS. As new features, this paper simulates the FDI attack and the syn flood DDoS attacks on 5G enabled remote Supervisory Control and Data Acquisition (SCADA) system that controls the solar photovoltaics (PV) controller, Battery Energy Storage (BES) controller, and EV controller of the EVCS. The extent of delay has been increased to more than 500 ms with the severe DDoS attack via 5G. The attacks make the EVCS system oscillate or shift the DC operating point. The frequency of oscillation, its damping, and the system's resiliency are found to be related to the attacks' intensity and the target controller. Finally, the novel stacked Long Short-Term Memory (LSTM) based intrusion detection systems (IDS) are proposed solely based on the electrical fingerprint. This model can detect the stealthy cyberattacks that bypass the cyber layer and go unnoticed in the monitoring system with nearly 100% detection accuracy.

show abstract

“…A single LSTM unit is much more complex than a traditional neural unit. It has four gates: input gate, output gate, forget gate, and cell gate [35]. The LSTM cell takes input feature

x_{t}

along with cell state

c_{t - 1}

and hidden state

h_{t - 1}

from previous LSTM units and outputs the current cell state

c_{t}

and hidden state

h_{t}

.…”

Section: Proposed Ids Methodologymentioning

confidence: 99%

Exploring cybersecurity issues in 5G enabled electric vehicle charging station with deep learning

Basnet

Ali

2021

IET Generation Trans & Dist

View full text Add to dashboard Cite

show abstract

“…Assim, a saída do deep autencoderé uma representação aprendida das relações maisúteis entre suas características de entrada. A pesquisa sobre deep autencoder para detecção de intrusão, embora com várias propostas, ainda está em seu inicio [Yan et al 2020]. Outras abordagens, buscam a redução da dimensionalidade dos dados [Li et al 2020], ou mesmo a detecção de outliers [Cheng et al 2021], visando como objetivo final melhorias na precisão da detecção de intrusão.…”

Section: Deep Autoencodersunclassified

“…No entanto, os autores assumem que os eventos podem ser rotulados conforme a necessidade, e visa apenas diminuir os custos computacionais, independentemente da quantidade necessária de instâncias rotuladas nas atualizações do modelo. Por outro lado, [Yan et al 2020] aplica uma série de codificadores automáticos para a tarefa de extração de características. A abordagem propostaé capaz de aumentar a precisão, enquanto a tarefa de atualização do modeloé negligenciada.…”

Section: Trabalhos Relacionadosunclassified

Um modelo de detecção de intrusão baseado em deep autoencoders e transfer learning

Santos

Viegas

Santin

2021

Anais Do XXI Simpósio Brasileiro De Segurança Da Informação E De Sistemas Computacionais (SBSeg 2021)

View full text Add to dashboard Cite

As técnicas de aprendizado de máquina para detecção de intrusão baseada na rede geralmente pressupõem que o tráfego da rede não muda com o tempo ou que as atualizações do modelo podem ser realizadas facilmente. Neste artigo, propomos um novo modelo de detecção de intrusão baseado em deep autoencoders e transfer learning para facilitar a atualização do modelo. Experimentos realizados mostraram que as abordagens na literatura são incapazes de lidar com mudanças de tráfego de rede ao longo do tempo. A abordagem proposta é capaz de melhorar a taxa de falso positivo em até 23,9%, e fornecer taxas de precisão semelhantes às técnicas tradicionais, exigindo apenas 22% de dados de treinamento e 28% dos custos computacionais.

show abstract

“…At present, most research on WSN intrusion detection uses traditional machine learning methods to analyze network traffic data. Due to the growth in both the network's size and its user base, the WSN network will generate high-dimensional traffic data, and the traditional machine learning approach would encounter issues like poor feature extraction and detection accuracy, which cannot meet such an application environment [4].…”

Section: Introductionmentioning

confidence: 99%

Intrusion Detection Model for Wireless Sensor Networks Based on MC-GRU

Jingjing

Tongyu

Zhang

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

A crucial line of defense for the security of wireless sensor network (WSN) is intrusion detection. This research offers a new MC-GRU WSN intrusion detection model based on convolutional neural networks (CNN) and gated recurrent unit (GRU) to solve the issues of low detection accuracy and poor real-time detection in existing WSN intrusion detection algorithms. MC-GRU uses multiple convolutions to extract network data traffic features and uses the high-level features output after convolution operations as input parameters of the GRU network, which strengthens the learning of spatial and time series features of traffic data and improves the detection performance of the model. The experiment results based on the WSN-DS dataset show that the overall detection accuracy of the four types of attack of black hole, gray hole, flooding, and scheduling and normal behaviors reaches 99.57%, and it is also better than the existing WSN intrusion detection algorithms in real-time performance and classification ability.

show abstract

A Network Intrusion Detection Method Based on Stacked Autoencoder and LSTM

Cited by 48 publications

References 18 publications

Exploring cybersecurity issues in 5G enabled electric vehicle charging station with deep learning

Exploring cybersecurity issues in 5G enabled electric vehicle charging station with deep learning

Um modelo de detecção de intrusão baseado em deep autoencoders e transfer learning

Intrusion Detection Model for Wireless Sensor Networks Based on MC-GRU

Contact Info

Product

Resources

About