Early Warning System for DDoS Attacking Based on Multilayer Deployment of Time Delay Neural Network

Tsai, Chen-Chi; Chang, Allen Y.; Huang, Ming-Szu

doi:10.1109/iihmsp.2010.178

Cited by 17 publications

(15 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…22,23 It is a network that has a hidden time factor inside the signal with implicit representation. 24 Its modeling process includes extracting information from the present and past inputs. 25 The difference between TDNN and ANN topologies is the existence of tapped delay line memory at the input to the first layer of the static feed-forward network in TDNN.…”

Section: Time Delay Neural Network Modelingmentioning

confidence: 99%

Modeling of occupant’s head movement behavior in motion sickness study via time delay neural network

et al. 2019

View full text Add to dashboard Cite

Passengers are more susceptible to experiencing motion sickness (MS) than drivers. The difference in the severity of MS is due to their different head movement behavior during curve driving. When negotiating a curve, the passengers tilt their heads towards the lateral acceleration direction while the drivers tilt their heads against it. Thus, to reduce the passengers’ level of MS, they need to reduce their head’s tilting angle towards the lateral acceleration direction. Designing MS minimization strategies is easier if the correlation between the head movement and lateral acceleration is known mathematically. Therefore, this paper proposes the utilization of a time delay neural network (TDNN) to model the correlation of the occupant’s head movement and lateral acceleration. An experiment was conducted to gather real-time data for the modeling process. The results show that TDNN manages to model the correlation by producing a similar output response to the actual response. Thus, it is expected that the correlation model could be used as an occupant’s head movement predictor tool in future studies of MS.

show abstract

Section: Time Delay Neural Network Modelingmentioning

confidence: 99%

Modeling of occupant’s head movement behavior in motion sickness study via time delay neural network

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Other studies have particularly tackled the problem of developing early warning techniques for DDoS attacks [7], [8], [9]. In [8], Xiao, Chen, and He proposed a cooperative system to produce warning signals.…”

Section: Related Workmentioning

confidence: 99%

“…The authors goal lies in reducing storage and computational resources consumption. In [7], Tsai, Chang and Huang presented a multi-layer system based on time delay neural networks. It is a cooperative system in which each device in the network monitors its neighborhood.…”

Section: Related Workmentioning

confidence: 99%

A Distributed Architecture for DDoS Prediction and Bot Detection

2020

View full text Add to dashboard Cite

Distributed Denial-of-Service (DDoS) attacks disrupt servers, services and the network, overloading the target resources and denying normal traffic. In order to defend from this attack, mitigation actions usually overprovision and sinkhole malicious traffic. Sooner the attack is detected, better is the mitigation. Hence, we advocate for using a prediction technique aiming to anticipate actions against the possible attack, before it effectively starts. Then, this article contributes to advance the state-of-the-art presenting a distributed architecture that identifies early signals of a possible DDoS attack and detects bots composing a botnet. The architecture identifies the malicious actors (bots) participating in the attack. The bot detection technique is triggered by the prediction of DDoS supported by early signals. Prediction identifies signals of attack on the network before it reaches advanced stages. Based on the metastability theory, it provides unsupervised statistical learning and identifies the imminence of DDoS attacks. The botnet detection is challenging because of the high dimension of data involved and because of resource constraints (memory and processing) in network devices. Network devices are clustered based on features extracted from the traffic and based on the causality between devices. Detection is performed per cluster. Performance evaluations took as input the CTU-13 Czech Republic University, CAIDA and Botnet 2014 datasets, efficiently detecting the bots in the dataset with an accuracy of 99.9%.

show abstract

“…T h ¼ Transfer Packet size =Inter arrival time attacker (11) i:e: T h ¼ Transfer Packet size Â frequency attacker…”

Section: Statistical Feature Of Clients Selectedmentioning

confidence: 99%

“…The use of time series models produces more errors when we make a wrong assumption of the distribution. An early warning system for DDoS attack detection is proposed using time delay neural networks [11]. The method supplements the working of traditional intrusion detection system (IDS).…”

Section: Introductionmentioning

confidence: 99%

Mathematical modelling of DDoS attack and detection using correlation

Singh

2017

Journal of Cyber Security Technology

View full text Add to dashboard Cite

Distributed denial of service attack (DDoS) is one of the top-rated cyber threats currently. It runs down the victim server resources such as bandwidth and buffer size by obstructing the server to provide resources to legitimate clients. In this article, we propose a mathematical model of DDoS attack; we discuss its relation to the features such as inter-arrival time or rate of arrival of the attacking clients accessing the server. We further analyse the attack model in context to exhausting bandwidth and buffer size of the victim server. The proposed method uses an unsupervised learning method, self-organising map, to form the clusters of identical features. Lastly, the article applies mathematical correlation and the normal probability distribution on the clusters and analyses their behaviours to detect a DDoS attack. The article uses the CAIDA 2007 data set for analysing and confirmation of the proposed model.

show abstract

Early Warning System for DDoS Attacking Based on Multilayer Deployment of Time Delay Neural Network

Cited by 17 publications

References 4 publications

Modeling of occupant’s head movement behavior in motion sickness study via time delay neural network

Modeling of occupant’s head movement behavior in motion sickness study via time delay neural network

A Distributed Architecture for DDoS Prediction and Bot Detection

Mathematical modelling of DDoS attack and detection using correlation

Contact Info

Product

Resources

About