Malicious Flows Research Articles

Botnet Attack Identification and Mitigation condition Software-Defined Networks Utilizing CNN Algorithm

One new design that makes managing and communicating across large-scale networks easier and more flexible is software-defined networking, or SDN. It allows for the smooth and dynamic execution of complicated network choices via programmable and centralized interfaces. But SDN opens doors for people and companies to tailor network apps to their needs, allowing them to enhance services. On the other hand, it began to encounter a host of new privacy and security issues and brought the dangers of one point of failure all at once. In most cases, hackers use OpenFlow switches to conduct botnets or distributed Denial of Service (DDoS) assaults against the controller. Popular security apps that use deep learning (DL) to quickly identify and counteract attacks are on the rise. Here, we examine botnet-based DDoS attack detection using DL approaches in an SDN-supported context and demonstrate their performance. For the assessment, we utilize a dataset that we just created ourselves. In order to choose the most useful subset of characteristics, we used weighting of features and tuning techniques. Using both a synthetic dataset and actual testbed conditions, we validate the measurements or simulation results. The primary objective of this research is to identify botnet-based DDoS assaults using easily-obtained characteristics and data using a lightweight DL approach with baseline hyper-parameters. We found that the DL technique's performance is affected by the optimal subset of features, and that the accuracy of predictions of the same approach may be varied with a different collection of features. Lastly, our empirical findings show that the CNN approach works better than both the dataset and the actual testbed environments. With CNN, the detection rate for typical flows is 99% and for malicious flows it drops to 97%.

Unveiling Malicious Network Flows Using Benford’s Law

The increasing proliferation of cyber-attacks threatening the security of computer networks has driven the development of more effective methods for identifying malicious network flows. The inclusion of statistical laws, such as Benford’s Law, and distance functions, applied to the first digits of network flow metadata, such as IP addresses or packet sizes, facilitates the detection of abnormal patterns in the digits. These techniques also allow for quantifying discrepancies between expected and suspicious flows, significantly enhancing the accuracy and speed of threat detection. This paper introduces a novel method for identifying and analyzing anomalies within computer networks. It integrates Benford’s Law into the analysis process and incorporates a range of distance functions, namely the Mean Absolute Deviation (MAD), the Kolmogorov–Smirnov test (KS), and the Kullback–Leibler divergence (KL), which serve as dispersion measures for quantifying the extent of anomalies detected in network flows. Benford’s Law is recognized for its effectiveness in identifying anomalous patterns, especially in detecting irregularities in the first digit of the data. In addition, Bayes’ Theorem was implemented in conjunction with the distance functions to enhance the detection of malicious traffic flows. Bayes’ Theorem provides a probabilistic perspective on whether a traffic flow is malicious or benign. This approach is characterized by its flexibility in incorporating new evidence, allowing the model to adapt to emerging malicious behavior patterns as they arise. Meanwhile, the distance functions offer a quantitative assessment, measuring specific differences between traffic flows, such as frequency, packet size, time between packets, and other relevant metadata. Integrating these techniques has increased the model’s sensitivity in detecting malicious flows, reducing the number of false positives and negatives, and enhancing the resolution and effectiveness of traffic analysis. Furthermore, these techniques expedite decisions regarding the nature of traffic flows based on a solid statistical foundation and provide a better understanding of the characteristics that define these flows, contributing to the comprehension of attack vectors and aiding in preventing future intrusions. The effectiveness and applicability of this joint method have been demonstrated through experiments with the CICIDS2017 public dataset, which was explicitly designed to simulate real scenarios and provide valuable information to security professionals when analyzing computer networks. The proposed methodology opens up new perspectives in investigating and detecting anomalies and intrusions in computer networks, which are often attributed to cyber-attacks. This development culminates in creating a promising model that stands out for its effectiveness and speed, accurately identifying possible intrusions with an F1 of nearly 80%, a recall of 99.42%, and an accuracy of 65.84%.

Introducing packet-level analysis in programmable data planes to advance Network Intrusion Detection

Programmable data planes offer precise control over the low-level processing steps applied to network packets, serving as a valuable tool for analysing malicious flows in the field of intrusion detection. Albeit with limitations on physical resources and capabilities, they allow for the efficient extraction of detailed traffic information, which can then be utilised by Machine Learning (ML) algorithms responsible for identifying security threats. In addressing resource constraints, existing solutions in the literature rely on compressing network data through the collection of statistical traffic features in the data plane. While this compression saves memory resources in switches and minimises the burden on the control channel between the data and the control plane, it also results in a loss of information available to the Network Intrusion Detection System (NIDS), limiting access to packet payload, categorical features, and the semantic understanding of network communications, such as the behaviour of packets within traffic flows. This paper proposes P4DDLe, a framework that exploits the flexibility of P4-based programmable data planes for packet-level feature extraction and pre-processing. P4DDLe leverages the programmable data plane to extract raw packet features from the network traffic, categorical features included, and to organise them in a way that the semantics of traffic flows are preserved. To minimise memory and control channel overheads, P4DDLe selectively processes and filters packet-level data, so that only the features required by the NIDS are collected. The experimental evaluation with recent Distributed Denial of Service (DDoS) attack data demonstrates that the proposed approach is very efficient in collecting compact and high-quality representations of network flows, ensuring precise detection of DDoS attacks.

Detecting and mitigating DDoS attacks with moving target defense approach based on automated flow classification in SDN networks

The Distributed Denial of Service (DDoS) coordinates synchronized attacks on systems on the Internet using a set of infected hosts (bots). Bots are programmed to attack a determined target by firing a lot of synchronized requests, causing slowness or unavailability of the service. This type of attack has recently grown in magnitude, diversity, and economic cost. Thus, this paper presents a DDoS detection and mitigation architecture based on Software Defined Networking (SDN). It considers the Moving Target Defense (MTD) approach, redirecting malicious floods to expendable low-capacity servers to protect the main server while discouraging the attacker. The redirecting decision is based on a sensor, that employs Machine Learning (ML) algorithms for flow classification. When malicious flows are detected, the sensor notifies the SDN controller to include them in the malicious hosts lists and to realize the redirection. The validation and evaluation of the proposed architecture are conducted by simulation. Results considering different classification models (probabilistic, linear model, neural networks, and trees) and attack types indicate that the proposed architecture is efficient in detecting and mitigating DDoS attacks in approximately 3 seconds.

Correlation of cyber threat intelligence with sightings for intelligence assessment and augmentation

Cyber threat intelligence (CTI) provides the means to rapidly identify and investigate attacks such that the security risks can be addressed. However, few studies have assessed the consistency between the CTI and the observations in the real-world environment (i.e., sightings). Accordingly, this study proposes an approach for assessing such consistency. The assessment process involves finding both false positives (i.e., attacks reported in the CTI, but not observed in the sightings) and false negatives (i.e., attacks observed in the sightings, but not reported in the CTI). The latter are then used to augment the CTI. Several strategies are proposed for assessment and augmentation with a large number of flows in the sightings. For assessment, we first list the characteristic rules for various attacks, and see whether the characteristics of the malicious flows labeled with the attack tags by the CTI match the corresponding rules. We also divide the reported malicious flows into clusters for easier observation. For augmentation, a machine learning framework is employed to identify flows in the sighting with a behavior similar to that of known malicious flows. The attack type and severity of these flows are predicted and used to update the CTI accordingly. The experimental results reveal that among the sightings, over 50% of the flows do not exhibit the behaviors expected from the characteristic rules, but nevertheless appear to be probing or scanning. The proportion of such flows is greater than 90% in the largest cluster for each attack type. When the learning framework is employed, the number of high-severity malicious sources identified in the sighting increases by 156% compared to that reported in the original blacklist. In addition, around 53% of these sources are also considered as potentially malicious by other intelligence sources, and are thus regarded as valid candidates for CTI augmentation.

A Feedback Semi-Supervised Learning With Meta-Gradient for Intrusion Detection

The widespread popularity of the Internet of human life has been accompanied by a significant increase in the cost of protecting private data from malicious attacks. Researchers have proposed many deep learning-based intrusion detection methods. However, traditional methods rely on a large number of unpolluted data to learn benign data distributions, while nonidentical distributions will affect the performance in distinguishing normal and abnormal data. To address this problem, this article proposes an intrusion detection method feedback semi-supervised learning with meta-gradient for intrusion detection (FSMG) based on feedback deep semi-supervised learning. FSMG constructs a lightweight evaluation network with slight data augmentation and nonprocessing on the same input, respectively, and uses a small amount of labeled data to infer nonidentically distributed data flows hidden in the training dataset. Then, FSMG converts malicious flows into useful information, continuously track and update the model, reducing data labeling errors. Furthermore, for the updating of gradients in the model, a bilevel nested optimization is used to ensure the model converges within <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$O({\mathrm{C}}/{\sqrt{T}})$</tex-math></inline-formula> . Unlike other semisupervised algorithms, FSMG uses labeled data and nonidentically distributed unlabeled data proportionally to construct the training dataset, achieving higher classification accuracy, and better robustness even with an 80% polluted rate.

Unsupervised Detection and Clustering of Malicious TLS Flows

Malware abuses TLS to encrypt its malicious traffic, preventing examination by content signatures and deep packet inspection. Network detection of malicious TLS flows is important, but it is a challenging problem. Prior works have proposed supervised machine learning detectors using TLS features. However, by trying to represent all malicious traffic, supervised binary detectors produce models that are too loose, thus introducing errors. Furthermore, they do not distinguish flows generated by different malware. On the other hand, supervised multiclass detectors produce tighter models and can classify flows by the malware family but require family labels, which are not available for many samples. To address these limitations, this work proposes a novel unsupervised approach to detect and cluster malicious TLS flows. Our approach takes input network traces from sandboxes. It clusters similar TLS flows using 90 features that capture properties of the TLS client, TLS server, certificate, and encrypted payload and uses the clusters to build an unsupervised detector that can assign a malicious flow to the cluster it belongs to, or determine if it is benign. We evaluate our approach using 972K traces from a commercial sandbox and 35M TLS flows from a research network. Our clustering shows very high precision and recall with an F1 score of 0.993. We compare our unsupervised detector with two state-of-the-art approaches, showing that it outperforms both. The false detection rate of our detector is 0.032% measured over four months of traffic.

Certrust: An SDN-Based Framework for the Trust of Certificates against Crossfire Attacks in IoT Scenarios

The low-intensity attack flows used by Crossfire attacks are hard to distinguish from legitimate flows. Traditional methods to identify the malicious flows in Crossfire attacks are rerouting, which is based on statistics. In these existing mechanisms, the identification of malicious flows depends on the IP address. However, the IP address is easy to be changed by attacks. Compared with the IP address, the certificate is more challenging to be tampered with or forged. Moreover, the traffic trend in the network is towards encryption. The certificates are popularly utilized by IoT devices for authentication in encryption protocols. DTLShps proposed a new way to verify certificates for resource-constrained IoT devices by using the SDN controller. Based on DTLShps, the SDN controller can collect statistics on certificates. In this paper, we propose Certrust, a framework based on the trust of certificates, to mitigate the Crossfire attack by using SDN for IoT. Our goal is threefold. First, the trust model is built based on the Bayesian trust system with the statistics on the participation of certificates in each Crossfire attack. Moreover, the forgetting curve is utilized instead of the traditional decay method in the Bayesian trust system for achieving a moderate decay rate. Second, for detecting the Crossfire attack accurately, a method based on graph connectivity is proposed. Third, several trust-based routing principles are proposed to mitigate the Crossfire attack. These principles can also encourage users to use certificates in communication. The performance evaluation shows that Certrust is more effective in mitigating the Crossfire attack than the traditional rerouting schemes. Moreover, our trust model has a more appropriate decay rate than the traditional methods.

LtRFT: Mitigate the Low-Rate Data Plane DDoS Attack With Learning-To-Rank Enabled Flow Tables

Software-Defined Networking (SDN) switches typically have limited ternary content addressable memory (TCAM) that caches the flow entries on the data plane. The scarcity and strong resource competitiveness of TCAM space put the flow tables at the risk of malicious Distributed Denial-of-Service (DDoS) attacks. In this paper, we propose LtRFT, a Learning-To-Rank (LtR) based scheme for mitigating the low-rate DDoS attacks targeted at flow tables. LtRFT consists of three modules: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">monitor</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ranker</i> , and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mitigator</i> . <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Monitor</i> manages the flow table status and sends alerts to other modules after detecting attacks. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Ranker</i> models the attack mitigation problem as a flow entry ranking task, and ranks malicious flows with a high eviction priority using a pairwise-based LtR algorithm. The <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mitigator</i> frees up the flow table space by deleting malicious flow entries according to the flow entry ranking sequence generated by <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ranker</i> . We introduce LtR to network attack detection innovatively and use both classification and information retrieval metrics to describe and evaluate LtRFT. Extensive experiments were conducted to validate the effectiveness and robustness of LtRFT in detecting and mitigating the low-rate data plane DDoS attacks. LtRFT can detect malicious attack flows with an accuracy of over 96%, and can reduce the attack flow duration by 97.7% with an average extra latency of 0.5 seconds, which proves that LtRFT is practicable in SDN deployments.

Twin-Node Neighbour Attack on AODV based Wireless Ad Hoc Network

As security is a very challenging issue in ad hoc networks, variety of research works related to security of ad hoc networks are being reported for last many years. In the present work, we propose a new sort of attack titled Twin-Node Neighbour Attack (TNNA), wherein two malicious nodes in close vicinity of each other exploits the provision of broadcast nature of Hello Messages in AODV routing protocol along with non-provision of any restriction regarding authentication of participating nodes. Mitigation measures are designed to lessen or perhaps remove security flaws and threats altogether. Detection and mitigation of TNNA attack are also proposed and discussed. The network's performance has been measured using four metrics viz. Packet Delivery Ratio, Throughput, Total Number of Received Packets and Average End-toEnd Delay. It is evident from simulations that the TNNA attack is significantly detrimental to the performance of WANETs using AODV routing protocol. After attack throughput of legitimate flow is found to be less than 5 % as compared to the Throughput without attack, when the data rate of malicious node is 100 Kibps. Due to stress of malicious flow of 100 Kibps, the number of transmitted (received) data packets of legitimate flow is reduced by a factor of more than 20.

Software defined network-based HTTP flooding attack defender

In recent years, the explosive growth of the Internet has led to an increment in the number of Distributed Denial of Service (DDoS) attacks. HTTP Flooding is a critical DDoS attack that targets HTTP servers to prohibit users from receiving HTTP services. Moreover, it saturates the link bandwidth and consumes network resources. Because the attack is launched at the application layer, it is difficult to defend against it using current countermeasures such as firewall or Intrusion Prevention System (IPS).In this paper, we propose SHFD, which leverages the Software-Defined Networking (SDN) paradigm to mitigate HTTP flooding attacks. We implement SHFD as a defender module on the SDN controller to detect and mitigate the attack in the first place. Experimental results gathered from Mininet confirm that SHFD brings a significant improvement of 13% in detection time and 29% in the number of blocked malicious flows compared to the state-of-the-art approaches.

Blockchain Empowered Federated Learning for Distributed Network Security Behaviour Knowledge Base in 6G

The malicious flow originating from massive access devices in 6G network will increase sharply. In order to effectively reduce malicious flow, we hope to establish a new framework for coordination of security monitoring and malicious behaviour control in 6G network. Federated learning provides data and privacy protection for the distributed network security behaviour knowledge base. However, since the equipment of its participants needs to upload the original data to the central server for model training, this may lead to data leakage in the knowledge base. Therefore, in this article, we first use the knowledge graph to describe network security behaviours, then build a universal network security malicious behaviour knowledge base, and discuss its application scenarios. Then, we propose a blockchain empowered federated learning (BeFL) for distributed network security malicious behaviour knowledge base architecture to ensure the security of knowledge transmission. Finally, we deployed the designed distributed knowledge base in the prototype system and compared it with the other two baseline methods to verify the performance. Relevant results show that our method outperforms other methods in terms of user identification, flow detection, and attack source tracing.

DEV-ETA: An Interpretable Detection Framework for Encrypted Malicious Traffic

Abstract Traffic encrypted technology enables Internet users to protect their data secrecy, but it also brings a challenge to malicious package detection. To tackle this issue, researchers have investigated into encrypted traffic analysis (ETA) in recent years. Existing works, however, only focus on the accuracy of malicious flow identification. Using ETA as a technical black box, they pay little attention to the internal details and explanation of models. In this paper, we, for the first time, introduce interpretable machine learning into ETA. We aim to provide a reasonable explanation for detection results, so as to enable one to understand and further trust network security analysts. We develop a complete analysis framework, named DEV-ETA (detection, explanation and verification of ETA). DEV-ETA applies post hoc interpretation methods to explain the detection results and verify the explanation using the joint distribution of support features on the dataset. We run thorough experiments to explain the detection result using three popular explanation approaches, namely SHAP, LIME and MSS, and we verify the explanation via the feature distribution plot. The experimental results show that our design can interpret the detection result of ETA model instead of just simply treating the model as a black box.

Network traffic analysis through deep learning for detection of an army of bots in health IoT network

PurposeIoT has a wide range of applications in the health-care sector and has captured the interest of many academic and industrial communities. The health IoT devices suffer from botnet attacks as all the devices are connected to the internet. An army of compromised bots may form to launch a DDoS attack, steal confidential data of patients and disrupt the service, and hence detecting this army of bots is paramount. This study aims to detect botnet attacks in health IoT devices using the deep learning technique.Design/methodology/approachThis paper focuses on designing a method to protect health IoT devices from botnet attacks by constantly observing communication network traffic and classifying them as benign and malicious flow. The proposed algorithm analyzes the health IoT network traffic through implementing Bidirectional long-short term memory, a deep learning technique. The IoT-23 data set is considered for this research as it includes diverse botnet attack scenarios.FindingsThe performance of the proposed method is evaluated using attack prediction accuracy. It results in the highest accuracy of 84.8%, classifying benign and malicious traffic.Originality/valueThe proposed method constantly monitors the health IoT network to detect botnet attacks and classifies the traffic as benign or attack. The system is implemented using the BiLSTM algorithm and trained using the IoT-23 data set. The diversity of attack scenarios of the IoT-23 data set demonstrates the proposed algorithm's competence in detecting botnet types in a heterogeneous environment.

Investigating the Effect of Traffic Sampling on Machine Learning-Based Network Intrusion Detection Approaches

Machine Learning (ML) based Network Intrusion Systems (NIDSs) operate on flow features which are obtained from flow exporting protocols (<i>i.e.,</i> NetFlow). Recent success of ML and Deep Learning (DL) based NIDS solutions assume such flow information (<i>e.g.,</i> avg. packet size) is obtained from all packets of the flow. However, often in practice flow exporter is deployed on commodity devices where packet sampling is inevitable. As a result, applicability of such ML based NIDS solutions in the presence of sampling (<i>i.e.,</i> when flow information is obtained from sampled set of packets instead of full traffic) is an open question. In this study, we explore the impact of packet sampling on the performance and efficiency of ML-based NIDSs. Unlike previous work, our proposed evaluation procedure is immune to different settings of flow export stage. Hence, it can provide a robust evaluation of NIDS even in the presence of sampling. Through sampling experiments we established that malicious flows with shorter size (<i>i.e.,</i> number of packets) are likely to go unnoticed even with mild sampling rates such as 1/10 and 1/100. Next, using the proposed evaluation procedure we investigated the impact of various sampling techniques on NIDS detection rate and false alarm rate. Detection rate and false alarm rate is computed for three sampling rates (<i>i.e.,</i> 1/10, 1/100, 1/1000), for four different sampling techniques and for three (two tree-based, one deep learning based) classifiers. Experimental results show that systematic linear sampler - SketFlow performs better compared to non-linear samplers such as Sketch Guided and Fast Filtered sampling. We also found that random forest classifier with SketchFlow sampling was a better combination. The combination showed higher detection rate and lower false alarm rate across multiple sampling rates compared to other sampler-classifier combinations. Our results are consistent in multiple sampling rates, exceptional case is observed for Sketch Guided Sampling (SGS) as it caused a drastic performance drop when sampling rate was changed from 1/100 to 1/1000. Our results provide valuable insights for network practitioners and researchers regarding on how packet sampling effects ML-based NIDS performance. In this regard full source code for sampling and ML experiments has been released: github.com/Jumabek/sampledFlowMeter and github.com/Jumabek/nids-with-sampling

A MACHINE LEARNING CLASSIFICATION APPROACH TO DETECT TLS-BASED MALWARE USING ENTROPY-BASED FLOW SET FEATURES

Transport Layer Security (TLS) based malware is one of the most hazardous malware types, as it relies on encryption to conceal connections. Due to the complexity of TLS traffic decryption, several anomaly-based detection studies have been conducted to detect TLS-based malware using different features and machine learning (ML) algorithms. However, most of these studies utilized flow features with no feature transformation or relied on inefficient flow feature transformations like frequency-based periodicity analysis and outliers percentage. This paper introduces TLSMalDetect, a TLS-based malware detection approach that integrates periodicity-independent entropy-based flow set (EFS) features generated by a flow feature transformation technique to solve flow feature utilization issues in related research. EFS features effectiveness was evaluated in two ways: (1) by comparing them to the corresponding outliers percentage and flow features using four feature importance methods, and (2) by analyzing classification performance with and without EFS features. Moreover, new Transmission Control Protocol features not explored in literature were incorporated into TLSMalDetect, and their contribution was assessed. This study’s results proved EFS features of the number of packets sent and received were superior to related outliers percentage and flow features and could remarkably increase the performance up to ~42% in the case of Support Vector Machine accuracy. Furthermore, using the basic features, TLSMalDetect achieved the highest accuracy of 93.69% by Naïve Bayes (NB) among the ML algorithms applied. Also, from a comparison view, TLSMalDetect’s Random Forest precision of 98.99% and NB recall of 92.91% exceeded the best relevant findings of previous studies. These comparative results demonstrated the TLSMalDetect’s ability to detect more malware flows out of total malicious flows than existing works. It could also generate more actual alerts from overall alerts than earlier research.Transport Layer Security (TLS) based malware is one of the most hazardous malware types, as it relies on encryption to conceal connections. Due to the complexity of TLS traffic decryption, several anomaly-based detection studies have been conducted to detect TLS-based malware using different features and machine learning (ML) algorithms. However, most of these studies utilized flow features with no feature transformation or relied on inefficient flow feature transformations like frequency-based periodicity analysis and outliers percentage. This paper introduces TLSMalDetect, a TLS-based malware detection approach that integrates periodicity-independent entropy-based flow set (EFS) features generated by a flow feature transformation technique to solve flow feature utilization issues in related research. EFS features effectiveness was evaluated in two ways: (1) by comparing them to the corresponding outliers percentage and flow features using four feature importance methods, and (2) by analyzing classification performance with and without EFS features. Moreover, new Transmission Control Protocol features not explored in literature were incorporated into TLSMalDetect, and their contribution was assessed. This study’s results proved EFS features of the number of packets sent and received were superior to related outliers percentage and flow features and could remarkably increase the performance up to ~42% in the case of Support Vector Machine accuracy. Furthermore, using the basic features, TLSMalDetect achieved the highest accuracy of 93.69% by Naïve Bayes (NB) among the ML algorithms applied. Also, from a comparison view, TLSMalDetect’s Random Forest precision of 98.99% and NB recall of 92.91% exceeded the best relevant findings of previous studies. These comparative results demonstrated the TLSMalDetect’s ability to detect more malware flows out of total malicious flows than existing works. It could also generate more actual alerts from overall alerts than earlier research.

Identification of Distributed Denial of Services Anomalies by Using Combination of Entropy and Sequential Probabilities Ratio Test Methods.

One of the most dangerous kinds of attacks affecting computers is a distributed denial of services (DDoS) attack. The main goal of this attack is to bring the targeted machine down and make their services unavailable to legal users. This can be accomplished mainly by directing many machines to send a very large number of packets toward the specified machine to consume its resources and stop it from working. We implemented a method using Java based on entropy and sequential probabilities ratio test (ESPRT) methods to identify malicious flows and their switch interfaces that aid them in passing through. Entropy (E) is the first technique, and the sequential probabilities ratio test (SPRT) is the second technique. The entropy method alone compares its results with a certain threshold in order to make a decision. The accuracy and F-scores for entropy results thus changed when the threshold values changed. Using both entropy and SPRT removed the uncertainty associated with the entropy threshold. The false positive rate was also reduced when combining both techniques. Entropy-based detection methods divide incoming traffic into groups of traffic that have the same size. The size of these groups is determined by a parameter called window size. The Defense Advanced Research Projects Agency (DARPA) 1998, DARPA2000, and Canadian Institute for Cybersecurity (CIC-DDoS2019) databases were used to evaluate the implementation of this method. The metric of a confusion matrix was used to compare the ESPRT results with the results of other methods. The accuracy and f-scores for the DARPA 1998 dataset were 0.995 and 0.997, respectively, for the ESPRT method when the window size was set at 50 and 75 packets. The detection rate of ESPRT for the same dataset was 0.995 when the window size was set to 10 packets. The average accuracy for the DARPA 2000 dataset for ESPRT was 0.905, and the detection rate was 0.929. Finally, ESPRT was scalable to a multiple domain topology application.

Online traffic classification for malicious flows using efficient machine learning techniques

The rapid network technology growth causing various network problems, attacks are becoming more sophisticated than defenses. In this paper, we proposed traffic classification by using machine learning technique, and statistical flow features such as five tuples for the training dataset. A rule-based system, Snort is used to identify the severe harmfulness data packets and reduce the training set dimensionality to a manageable size. Comparison of performance between training dataset that consists of all priorities malicious flows with only has priority 1 malicious flows are done. Different machine learning (ML) algorithms performance in terms of accuracy and efficiency are analyzed. Results show that Naïve Bayes achieved accuracy up to 99.82% for all priorities while 99.92% for extracted priority 1 of malicious flows training dataset in 0.06 seconds and be chosen to classify traffic in real-time process. It is demonstrated that by taking just five tuples information as features and using Snort alert information to extract only important flows and reduce size of dataset is actually comprehensive enough to supply a classifier with high efficiency and accuracy which can sustain the safety of network.

Malicious traffic detection combined deep neural network with hierarchical attention mechanism

Given the gradual intensification of the current network security situation, malicious attack traffic is flooding the entire network environment, and the current malicious traffic detection model is insufficient in detection efficiency and detection performance. This paper proposes a data processing method that divides the flow data into data flow segments so that the model can improve the throughput per unit time to meet its detection efficiency. For this kind of data, a malicious traffic detection model with a hierarchical attention mechanism is also proposed and named HAGRU (Hierarchical Attention Gated Recurrent Unit). By fusing the feature information of the three hierarchies, the detection ability of the model is improved. An attention mechanism is introduced to focus on malicious flows in the data flow segment, which can reasonably utilize limited computing resources. Finally, compare the proposed model with the current state of the method on the datasets. The experimental results show that: the novel model performs well in different evaluation indicators (detection rate, false-positive rate, F-score), and it can improve the performance of category recognition with fewer samples when the data is unbalanced. At the same time, the training of the novel model on larger datasets will enhance the generalization ability and reduce the false alarm rate. The proposed model not only improves the performance of malicious traffic detection but also provides a new research method for improving the efficiency of model detection.

Detection and Classification of Malicious Flows in Software-Defined Networks Using Data Mining Techniques.

The increasing availability of mobile devices and applications, the progress in virtualisation technologies, and advances in the development of cloud-based distributed data centres have significantly stimulated the growing interest in the use of software-defined networks (SDNs) for both wired and wireless applications. Standards-based software abstraction between the network control plane and the underlying data forwarding plane, including both physical and virtual devices, provides an opportunity to significantly increase network security. In this paper, to secure SDNs against intruders’ actions, we propose a comprehensive system that exploits the advantages of SDNs’ native features and implements data mining to detect and classify malicious flows in the SDN data plane. The architecture of the system and its mechanisms are described, with an emphasis on flow rule generation and flow classification. The concept was verified in the SDN testbed environment that reflects typical SDN flows. The experiments confirmed that the system can be successfully implemented in SDNs to mitigate threats caused by different malicious activities of intruders. The results show that our combination of data mining techniques provides better detection and classification of malicious flows than other solutions.