Intrusion Detection of UAVs Based on the Deep Belief Network Optimized by PSO

Tan, Xiaopeng; Su, Shaojing; Zuo, Zhen; Guo, Xiaojun; Sun, Xiang

doi:10.3390/s19245529

Cited by 49 publications

(17 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…T . For each particle in the population, fitness value can be obtained according to fitness function to evaluate the fitness of particles [22]. This paper adopts the fitness function proposed by Vieira [23], as shown in (1), where α is a hyperparameter, which can coordinate the relationship between the classifier performance P (including accuracy, F-score, precision, etc.)…”

Section: Background a Partical Swarm Optimization (Pso)mentioning

confidence: 99%

Research on Intrusion Detection Based on Particle Swarm Optimization in IoT

et al. 2021

View full text Add to dashboard Cite

With the advent of the "Internet plus" era, the Internet of Things (IoT) is gradually penetrating into various fields, and the scale of its equipment is also showing an explosive growth trend. The age of the "Internet of Everything" is coming. The integration and diversification of IoT terminals and applications make IoT more vulnerable to various intrusion attacks. Therefore, it is particularly important to design an intrusion detection model that guarantees the security, integrity and reliability of the IoT. Traditional intrusion detection technology has the disadvantages of low detection rate and poor scalability, which cannot adapt to the complex and changeable IoT environment. In this paper, we propose a particle swarm optimization-based gradient descent (PSO-LightGBM) for the intrusion detection. In this method, PSO-LightGBM is used to extract the features of the data and inputs it into one-class SVM (OCSVM) to discover and identify malicious data. The UNSW-NB15 dataset is applied to verify the intrusion detection model. The experimental results show that the model we propose is very robust in detecting either normal or various malicious data, especially small sample data such as Backdoor, Shellcode and Worms. INDEX TERMS Intrusion detection, Internet of Things (IoT), particle swarm optimization (PSO), oneclass SVM(OCSVM)

show abstract

Section: Background a Partical Swarm Optimization (Pso)mentioning

confidence: 99%

Research on Intrusion Detection Based on Particle Swarm Optimization in IoT

et al. 2021

View full text Add to dashboard Cite

show abstract

“…PSO and its variants are widely used in network intrusion detection. Tan et al [25] proposed to use PSO algorithm to optimize deep belief network (DBN) and apply it to network intrusion detection. Their results showed that the effect of PSO was superior to intelligent optimization algorithms such as genetic algorithm (GA) and simulated annealing (SA), thus the PSO-DBN model also performed other machine learning models.…”

Section: Related Workmentioning

confidence: 99%

“…However, in [25], [26], the results only gave the overall metrics of the algorithm on the dataset, including accuracy, fmeasures and so on, but not the performance of the algorithm in each class. In network intrusion detection, common types of attacks are easier to detect, such as Dos attacks; while rare types of attacks, such as R2L and U2R, are more difficult to detect, because the system does not have enough database about the features of these attacks.…”

Section: Related Workmentioning

confidence: 99%

Network Intrusion Detection Based on PSO-Xgboost Model

Jiang

et al. 2020

IEEE Access

156

View full text Add to dashboard Cite

Network intrusion detection system (NIDS) is a commonly used tool to detect attacks and protect networks, while one of its general limitations is the false positive issue. On the basis of our comparative experiments and analysis for the characteristics of the particle swarm optimization (PSO) and Xgboost, this paper proposes the PSO-Xgboost model given its overall higher classification accuracy than other alternative models such like Xgboost, Random Forest, Bagging and Adaboost. Firstly, a classification model based on Xgboost is constructed, and then PSO is used to adaptively search for the optimal structure of Xgboost. The benchmark NSL-KDD dataset is used to evaluate the proposed model. Our experimental results demonstrate that PSO-Xgboost model outperforms other comparative models in precision, recall, macro-average (macro) and mean average precision (mAP), especially when identifying minority groups of attacks like U2R and R2L. This work also provides experimental arguments for the application of swarm intelligence in NIDS.

show abstract

“…The machine learning (ML) algorithms mentioned above depend on complex feature engineering and are difficult to adapt to the growing network environment, while the deep learning (DL) algorithm can autonomously abstract high-level features from basic network traffic without complex feature engineering; therefore, it is widely used in attack recognition tasks. Tan et al [18] obtained a structure-optimal deep belief network (DBN) optimized by the particle swarm optimization (PSO) method, and experiments showed that the accuracy of PSO-DBN on the KDD99 dataset reached 92.44%. Marir et al [19] applied DBN as a feature selector and combined it with an SVM classifier to improve the recognition accuracy of IDS.…”

Section: Introductionmentioning

confidence: 99%

A Novel Wireless Network Intrusion Detection Method Based on Adaptive Synthetic Sampling and an Improved Convolutional Neural Network

Wang

Lei

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The diversity of network attacks poses severe challenges to intrusion detection systems (IDSs). Traditional attack recognition methods usually adopt mining data associations to identify anomalies, which has the disadvantages of a high false alarm rate (FAR), low recognition accuracy (ACC) and poor generalization ability. To ameliorate the comprehensive capabilities of IDS and strengthen network security, we propose a novel intrusion detection method based on the adaptive synthetic sampling (ADASYN) algorithm and an improved convolutional neural network (CNN). First, we use the ADASYN method to balance the sample distribution, which can effectively prevent the model from being sensitive to large samples and ignore small samples. Second, the improved CNN is based on the split convolution module (SPC-CNN), which can increase the diversity of features and eliminate the impact of interchannel information redundancy on model training. Then, an AS-CNN model mixed with ADASYN and SPC-CNN is used for intrusion detection tasks. Finally, the standard NSL-KDD dataset is selected to test AS-CNN. The simulation illustrates that the accuracy is 4.60% and 2.79% higher than that of the traditional CNN and RNN models, and the detection rate (DR) increased by 11.34% and 10.27%, respectively. Additionally, the FAR decreased by 15.58% and 14.57%, respectively, compared with the two models.

show abstract

Intrusion Detection of UAVs Based on the Deep Belief Network Optimized by PSO

Cited by 49 publications

References 46 publications

Research on Intrusion Detection Based on Particle Swarm Optimization in IoT

Research on Intrusion Detection Based on Particle Swarm Optimization in IoT

Network Intrusion Detection Based on PSO-Xgboost Model

A Novel Wireless Network Intrusion Detection Method Based on Adaptive Synthetic Sampling and an Improved Convolutional Neural Network

Contact Info

Product

Resources

About