MF-CNN: a New Approach for LDoS Attack Detection Based on Multi-feature Fusion and CNN

Tang, Dan; Liu, Tang; Shi, Wei; Zhan, Sijia; Yang, Qiuwei

doi:10.1007/s11036-019-01506-1

Cited by 32 publications

(11 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It uses a hybrid intrusion detection system (IDS) to detect IoT-DoS attacks, which detects suspicious network traffic from network nodes based on long short-term memory (LSTM). In [25], [26] the application of deep learning in the detection and mitigation of DDoS attacks on SDN controllers using models such as LSTM and convolutional neural network (CNN) is discussed. The models were implemented to detect transmission control protocol (TCP), user datagram protocol (UDP), and internet control message protocol (ICMP) flood attacks.…”

Section: Literature Surveymentioning

confidence: 99%

An intelligent system to detect slow denial of service attacks in software-defined networks

John

Nagappasetty

2023

IJECE

View full text Add to dashboard Cite

<span lang="EN-US">Slow denial of service attack (DoS) is a tricky issue in software-defined network (SDN) as it uses less bandwidth to attack a server. In this paper, a slow-rate DoS attack called Slowloris is detected and mitigated on Apache2 and Nginx servers using a methodology called an intelligent system for slow DoS detection using machine learning (ISSDM) in SDN. Data generation module of ISSDM generates dataset with response time, the number of connections, timeout, and pattern match as features. Data are generated in a real environment using Apache2, Nginx server, Zodiac FX OpenFlow switch and Ryu controller. Monte Carlo simulation is used to estimate threshold values for attack classification. Further, ISSDM performs header inspection using regular expressions to mark flows as legitimate or attacked during data generation. The proposed feature selection module of ISSDM, called blended statistical and information gain (BSIG), selects those features that contribute best to classification. These features are used for classification by various machine learning and deep learning models. Results are compared with feature selection methods like Chi-square, T-test, and information gain.</span>

show abstract

Section: Literature Surveymentioning

confidence: 99%

An intelligent system to detect slow denial of service attacks in software-defined networks

John

Nagappasetty

2023

IJECE

View full text Add to dashboard Cite

show abstract

“…This method uses the Adaboost classification algorithm in the machine learning field to effectively identify LDDoS attack flow. Tang et al [23] proposed an LDDoS attack detection method based on multifeature fusion and Convolutional Neural Network (CNN) by calculating various network features and fusing them into feature maps. Siracusano et al [24] proposed an LDDoS attack detection method based on malicious TCP flow characteristics.…”

Section: Related Workmentioning

confidence: 99%

Extracting Low‐Rate DDoS Attack Characteristics: The Case of Multipath TCP‐Based Communication Networks

Lei

et al. 2021

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

The multipath TCP (MPTCP) enables multihomed mobile devices to realize multipath parallel transmission, which greatly improves the transmission performance of the mobile communication network. With the rapid development of all kinds of emerging technologies, network attacks have shown a trend of development with many types and rapid updates. Among them, low-rate distributed denial of service (LDDoS) attacks are considered to be one of the most threatening issues in the field of network security. In view of the current research status, by using the network simulation software NS2, this paper first compares and analyzes the throughput and delay performance of the MPTCP transmission system under LDDoS attacks and, further, conducts simulation experiments and analysis on the queue occupancy rate of the LDDoS attack flow to extract the basic attack characteristics of the LDDoS attacks. The experimental results show that the LDDoS attacks will have a major destructive effect on the throughput performance and delay performance of the MPTCP transmission system, resulting in a decrease in the robustness of the transmission system. By analyzing and comparing the occupancy rate of the LDDoS attack flow in the MPTCP transmission system, it can be concluded that (1) the occupancy rate of the LDDoS scattered pulse traffic sent by each puppet machine changes slightly, and (2) the occupancy rate of LDDoS attack data flow is much greater than that of ordinary TCP data flow.

show abstract

“…The DDoS assault may appear as a bottleneck that is created by the bridge and blocks normal traffic at its desired location. In [22], researchers presented an algorithm to improve the efficiency and robustness of the low-rate DoS (DDoS) detection system by combining PSDentropy and Support Vector Machine (SVM). Entropy application efficiently reduces the number of calculations, while SVM supports the proposed method by organizing the dataset around its most pertinent characteristics.…”

Section: Related Workmentioning

confidence: 99%

Multilayer Self-defense System to Protect Enterprise Cloud

Mishra¹,

Sharma²,

Alowaidi³

2020

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

A data breach can seriously impact organizational intellectual property, resources, time, and product value. The risk of system intrusion is augmented by the intrinsic openness of commonly utilized technologies like TCP/IP protocols. As TCP relies on IP addresses, an attacker may easily trace the IP address of the organization. Given that many organizations run the risk of data breach and cyber-attacks at a certain point, a repeatable and well-developed incident response framework is critical to shield them. Enterprise cloud possesses the challenges of security, lack of transparency, trust and loss of controls. Technology eases quickens the processing of information but holds numerous risks including hacking and confidentiality problems. The risk increases when the organization outsources the cloud storage services through the vendor and suffers from security breaches and need to create security systems to prevent data networks from being compromised. The business model also leads to insecurity issues which derail its popularity. An attack mitigation system is the best solution to protect online services from emerging cyber-attacks. This research focuses on cloud computing security, cyber threats, machine learning-based attack detection, and mitigation system. The proposed SDN-based multilayer machine learning-based self-defense system effectively detects and mitigates the cyber-attack and protects cloud-based enterprise solutions. The results show the accuracy of the proposed machine learning techniques and the effectiveness of attack detection and the mitigation system.

show abstract

MF-CNN: a New Approach for LDoS Attack Detection Based on Multi-feature Fusion and CNN

Cited by 32 publications

References 18 publications

An intelligent system to detect slow denial of service attacks in software-defined networks

An intelligent system to detect slow denial of service attacks in software-defined networks

Extracting Low‐Rate DDoS Attack Characteristics: The Case of Multipath TCP‐Based Communication Networks

Multilayer Self-defense System to Protect Enterprise Cloud

Contact Info

Product

Resources

About