Estimating Strength of a DDoS Attack in Real Time Using ANN Based Scheme

Agrawal, P. K.; Gupta, Brij B.; Jain, Satbir; Pattanshetti, Mahantesh K.

doi:10.1007/978-3-642-22786-8_38

Cited by 9 publications

(7 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural Networks [57,59,60,63] Magnitude of attack (number of zombies and attack rate) identification through LVQ model of neural network.…”

Section: Basis Of Defense Methodsmentioning

confidence: 99%

“…In other research attempts found in [59] and [60], authors use BP model of neural networks to estimate the strength of DDoS attack in real time and predict the number of zombies respectively. Backpropagation neural network is a multilayer feed forward network with backpropagation (feedback) of an error function [61].…”

Section: Application Of Neural Network In Ddos Detectionmentioning

confidence: 99%

“…This process is continued until network's output error is brought down to an acceptable level or the preset time of learning is achieved [62]. In [59], authors train the BP neural network with a dataset of variations in traffic entropy as inputs and the corresponding actual DDoS strengths as outputs. 20 different samples in the dataset are used for training with 10 Mbps attack strength as the lowest and 100 Mbps being the highest in the dataset.…”

Section: Application Of Neural Network In Ddos Detectionmentioning

confidence: 99%

See 2 more Smart Citations

A Survey on DDoS Attack and Defense Strategies: From Traditional Schemes to Current Techniques

Aamir

Zaidi

2013

IIS

View full text Add to dashboard Cite

Distributed Denial of Service (DDoS) attacks exhaust victim's bandwidth or services. Traditional architecture of Internet is vulnerable to DDoS attacks and an ongoing cycle of attack & defense is observed. A recent attack report of year 2013 -'Quarter 1' from Prolexic Technologies identifies that 1.75 percent increase in total number of DDoS attacks has been recorded as compared to similar attacks of previous year's last quarter. In this paper, different types and techniques of DDoS attacks and their countermeasures are surveyed. The significance of this paper is the coverage of many aspects of countering DDoS attacks including new research on the topic. We survey different papers describing methods of defense against DDoS attacks based on entropy variations, traffic anomaly parameters, neural networks, device level defense, botnet flux identifications, application layer DDoS defense and countermeasures in wireless networks, CCN & cloud computing environments. We also discuss some traditional methods of defense such as traceback and packet filtering techniques, so that readers can identify major differences between traditional and current techniques of defense against DDoS attacks. We identify that application layer DDoS attacks possess the ability to produce greater impact on the victim as they are driven by legitimate-like traffic, making it quite difficult to identify and distinguish from legitimate requests. The need of improved defense against such attacks is therefore more demanding in research. The study conducted in this paper can be helpful for readers and researchers to recognize better techniques of defense in current times against DDoS attacks and contribute with more research on this topic in the light of future challenges identified in this paper.

show abstract

“…Neural Networks [57,59,60,63] Magnitude of attack (number of zombies and attack rate) identification through LVQ model of neural network.…”

Section: Basis Of Defense Methodsmentioning

confidence: 99%

Section: Application Of Neural Network In Ddos Detectionmentioning

confidence: 99%

Section: Application Of Neural Network In Ddos Detectionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey on DDoS Attack and Defense Strategies: From Traditional Schemes to Current Techniques

Aamir

Zaidi

2013

IIS

View full text Add to dashboard Cite

show abstract

“…Synaptic connections are the carriers of the virtual signals which pass from one neuron network to another. The inter‐connections are modified by the weight vector whose values are adjusted according to the learning procedure 85 …”

Section: Dataset Descriptionmentioning

confidence: 99%

“…The inter-connections are modified by the weight vector whose values are adjusted according to the learning procedure. 85 The neural network is a cascaded structure with multiple layers; more layers are more efficient. Our ANN DDoS detection network is based on both supervised and unsupervised learning.…”

Section: Annmentioning

confidence: 99%

Performance evaluation of machine learning models for distributed denial of service attack detection using improved feature selection and hyper‐parameter optimization techniques

Habib

Khursheed

2022

Concurrency and Computation

View full text Add to dashboard Cite

This article gives the framework of extensive experimentation of various machine learning models to detect distributed denial of service attacks (DDoS). We use six-tier feature ranking methods that use statistical techniques as well as machine learning based classifiers to obtain the significant features. The measurable statistical based feature selection involves Chi-Square (Chi2), information gain (IG), merged Chi-Square (Chi2)-IG ranking and machine learning classifiers involve ensemble classifiers, that is, decision tree, random forest and eXtreme gradient boosting (XGBoost). Different supervised machine learning models (logistic regression, decision tree classifier, linear support vector machine, k-nearest neighbors, Gaussian Naive Bayes, random forest classifier, XGBoost) are trained on a feature-engineered datasets. To further our research, we use neural networks (ANN and CNN) using both feature-selected and auto-feature selection training setup. We check the validation and adaptability of these models with the optimal tuning of various parameters using GridSearchCV and the effectiveness of random sampling in overcoming the class imbalance problem. Based on various feature selection methods, the models are evaluated for their best performance. The experimental results show that our trained machine learning models and neural networks outperformed the ones in the state of art. The performance analysis is done based on confusion matrix scores, that is, accuracy, false alarm rate, sensitivity, specificity, false-positive rate, F1 score, area under curve analysis and loss functions on well-known KDD Cup 99 and UNSW-NB15 datasets. This study is significant for furthering the research in DDoS detection with machine learning and deep neural networks.

show abstract