Filters








25,791 Hits in 3.9 sec

Flow-Packet Hybrid Traffic Classification for Class-Aware Network Routing [article]

Sayantan Chowdhury, Ben Liang, Ali Tizghadam, Ilijc Albanese
2021 arXiv   pre-print
At a network router, the packets need to be processed with minimum delay, so the classifier cannot wait until the end of the flow to make a decision.  ...  from a flow-based classifier residing outside the router.  ...  % Flow-based Classifier with LightGBM Flow-based Classifier with XGBoost FPHTC with LightGBM FPHTC with XGBoost Regular packet-based traffic classification Table 2 : 2 Balanced test accuracy  ... 
arXiv:2105.00074v1 fatcat:rlqxqvhyyfhgbnlcbmyautfdw4

On Internet Traffic Classification: A Two-Phased Machine Learning Approach

Taimur Bakhshi, Bogdan Ghita
2016 Journal of Computer Networks and Communications  
The resulting classifier reported an average accuracy of 92.37% on approximately 3.4 million test cases increasing to 96.67% with adaptive boosting.  ...  Furthermore, the computational performance and accuracy of the proposed methodology in comparison with similar machine learning techniques lead us to recommend its extension to other applications in achieving  ...  In terms of overall accuracy, twophased ML provided a much more coherent and applicable result at 96.67% with the lowest accuracy attributed to SMO at approximately 53.2% correctly classified records.  ... 
doi:10.1155/2016/2048302 fatcat:swsnptehsvhrrbsswo24jdj244

Flow-Aware Elephant Flow Detection for Software-Defined Networks

Mosab Hamdan, Bushra Mohammed, Usman Humayun, Ahmed Abdelaziz, Suleman Khan, M Akhtar Ali, Muhammad Imran, MN Marsono.
2020 IEEE Access  
The proposed technique employs two classifiers, each respectively on SDN switches and controller, to achieve accurate elephant flow detection efficiently.  ...  Software-defined networking (SDN) separates the network control plane from the packet forwarding plane, which provides comprehensive network-state visibility for better network management and resilience  ...  We then compare it with other methods in terms of accuracy, precision, recall, F-measure, and running time. 1) CLASSIFICATION ACCURACY Accuracy is one of the essential classifiers metrics.  ... 
doi:10.1109/access.2020.2987977 fatcat:4l7xkadzwra5bbkh6lbu6qmuqi

Detecting IoT Botnet in 5G Core Network Using Machine Learning

Ye-Eun Kim, Min-Gyu Kim, Hwankuk Kim
2022 Computers Materials & Continua  
In both classification methods, the IoT Botnet detection performance using only 5GC's GTP-U packets decreased by at least 22.99% of accuracy compared to detection in wired network environment.  ...  As Internet of Things (IoT) devices with security issues are connected to 5G mobile networks, the importance of IoT Botnet detection research in mobile network environments is increasing.  ...  The gNB provides the UE with the protocols for the Control and User Planes, General Packet Radio Service Tunneling Protocol (GTP).  ... 
doi:10.32604/cmc.2022.026581 fatcat:jxfj4jihabg77ovzawyhrebwam

Wearable networked sensing for human mobility and activity analytics: A systems study

Bo Dong, Subir Biswas
2012 2012 Fourth International Conference on Communication Systems and Networks (COMSNETS 2012)  
Specific machine learning mechanisms are implemented for recognizing a target set of activities with both out-of-body and on-body processing arrangements.  ...  Impacts of energy consumption by the on-body sensors are analyzed in terms of activity detection accuracy for out-of-body processing.  ...  A packet occupies approximately 30ms, and the allocated guard time is approximately 20ms. Fig. 2 :b depicts the polling and data packet structures.  ... 
doi:10.1109/comsnets.2012.6151376 pmid:25530911 pmcid:PMC4269838 dblp:conf/comsnets/DongB12 fatcat:774plant7rcslmcphfdrd4hywa

An Evolutionary Algorithm for Enhanced Magnetic Resonance Imaging Classification

T.S. Murunya, S. Audithan
2014 Research Journal of Applied Sciences Engineering and Technology  
The proposed framework extracts features using Moment Invariants (MI) and Wavelet Packet Tree (WPT).  ...  Naïve Bayes and K-Nearest Neighbor (KNN) classify the selected features. National Biomedical Imaging Archive (NBIA) dataset including colon, brain and chest is used to evaluate the framework.  ...  Hence, wavelet packet analysis provides better frequency resolution control for signal decomposition (Chang and Kuo, 1993) .  ... 
doi:10.19026/rjaset.8.1205 fatcat:a5ni43eazjdffnu7vn3rbaozom

Semi-supervised Learning Framework for UAV Detection [article]

Olusiji O Medaiyese, Martins Ezuma, Adrian P Lauf, Ismail Guvenc
2021 arXiv   pre-print
When detecting the presence of RF-based UAV, we achieved an accuracy of 96.7% and 86% at a signal-to-noise ratio of 30 dB and 18 dB, respectively.  ...  By decomposing the RF signals using a two-level wavelet packet transform, we estimated the second moment statistic (i.e., variance) of the coefficients in each packet as a feature set.  ...  The feature set is used to train a classifier for the novel detection of UAV control signals.  ... 
arXiv:2104.06614v1 fatcat:mzi3lzwqgjgk5ew33zxnnzicim

Classification of the mechanomyogram signal using a wavelet packet transform and singular value decomposition for multifunction prosthesis control

Hong-Bo Xie, Yong-Ping Zheng, Jing-Yi Guo
2009 Physiological Measurement  
Instead of dividing only the approximation spaces, wavelet packet bases present both approximation and detail spaces in a binary tree by recursive splitting of vector spaces.  ...  So, it is not approximate to use the time domain feature, i.e. mean absolute value (MAV), zero crossing (ZC), etc, as a pattern classifier input, which gave great success in EMG prosthetic control (Hudgins  ... 
doi:10.1088/0967-3334/30/5/002 pmid:19349648 fatcat:yx3e3p6lcjfflnlm5mriz7cxzy

Enhancing congestion control to address link failure loss over mobile ad-hoc network [article]

Mohammad Amin Kheirandish Fard, Sasan Karamizadeh, Mohammad Aflaki
2011 arXiv   pre-print
The importance of detecting and responding link failure losses is to prevent sender from remaining idle unnecessarily and manage number of packet retransmission overhead.  ...  Standard congestion control cannot detect link failure losses which occur due to mobility and power scarcity in multi-hop Ad-Hoc network (MANET).  ...  The packet size is 1000 bytes and the routing protocol used in Not only accuracy of classifying loss due to link failure should be satisfactory, but also accuracy of classifying loss due to congestion  ... 
arXiv:1110.2289v1 fatcat:noe5uv44lvcufhqm5crqys5ijy

Enhancing Congestion Control To Address Link Failure Loss Over Mobile Ad-Hoc Network

Mohammad Amin Kheirandish Fard, Sasan Karamizadeh, Mohammad Aflaki
2011 International Journal of Computer Networks & Communications  
The importance of detecting and responding link failure losses is to prevent sender from remaining idle unnecessarily and manage number of packet retransmission overhead.  ...  Standard congestion control cannot detect link failure losses which occur due to mobility and power scarcity in multi-hop Ad-Hoc network (MANET).  ...  The packet size is 1000 bytes and the routing protocol used in Not only accuracy of classifying loss due to link failure should be satisfactory, but also accuracy of classifying loss due to congestion  ... 
doi:10.5121/ijcnc.2011.3513 fatcat:ptbyw6xxmrhu3dqaxuzzit6znm

HDLIDP: A Hybrid Deep Learning Intrusion Detection and Prevention Framework

Magdy M. Fadel, Sally M. El-Ghamrawy, Amr M. T. Ali-Eldin, Mohammed K. Hassan, Ali I. El-Desoky
2022 Computers Materials & Continua  
This framework improves detection accuracy while addressing all of the aforementioned problems.  ...  To validate the framework, experiments are done on both traditional and SDN datasets; the findings demonstrate a significant improvement in classification accuracy.  ...  According to the results obtained, it is advised to deploy the framework with two hidden layers NN as it has best accuracy and approximately good running time.  ... 
doi:10.32604/cmc.2022.028287 fatcat:ppfmbcq5zffyzpuw6vk3ihyce4

Minimizing false positive rate for DoS attack detection: A hybrid SDN-based approach

Majd Latah, Levent Toker
2019 ICT Express  
Our system combines flow-based and packet-based approaches to minimize the false positive rate (FPR).  ...  Recently, a novel networking paradigm that provides enhanced programming abilities has been proposed to attain an efficient control and management in future networks.  ...  Finally, combining kNN approach as a packet-based along with neural networks approach as a flowbased approach shows the highest accuracy, minimum false positive rate and the highest precision when compared  ... 
doi:10.1016/j.icte.2019.11.002 fatcat:evy6v5bumrgj7mommghxymsmxy

Towards Detecting Flooding DDOS Attacks Over Software Defined Networks Using Machine Learning Techniques

Ancy Sherin Jose, Latha R Nair, Varghese Paul
2021 Revista GEINTEC  
In this case, the controller of the SDN gets overloaded with the incoming packets from the switches.  ...  The system uses various classification algorithms to classify a traffic into normal or attack.  ...  Performance of classifiers with feature group 2 with 4 selected features, achieve highest accuracy of 99.73% for SVM classifier.  ... 
doi:10.47059/revistageintec.v11i4.2411 fatcat:xvb3jj7n4jc4vawwy6wcpiyy3q

Privacy Failures in Encrypted Messaging Services: Apple iMessage and Beyond [article]

Scott Coull, Kevin Dyer
2014 arXiv   pre-print
In this paper, however, we show that it is possible for an eavesdropper to learn information about user actions, the language of messages, and even the length of those messages with greater than 96% accuracy  ...  despite the use of state-of-the-art encryption technologies simply by observing the sizes of encrypted packet.  ...  The results indicate that we are able to accurately classify the OS with 100% accuracy after observing only five packets regardless of the operating system.  ... 
arXiv:1403.1906v1 fatcat:huuydpibjrespbj5reh257krdu

Speech Emotion Recognition in Acted and Spontaneous Context

Farah Chenchah, Zied Lachiri
2014 Procedia Computer Science  
The experimental results shows that ERB scale features gives better performance in comparison with other studied features with recognition accuracy of 78.75% for acted context and 50.06% for spontaneous  ...  For the purpose of this work, we have examined wavelet packet energy and entropy features applied to Mel, Bark and ERB scale applied with Hidden Markov Model (HMM) as classification system.  ...  The results show that wavelet packet filter bank with ERB scale give promising classification accuracy for both of databases.  ... 
doi:10.1016/j.procs.2014.11.020 fatcat:zj6vrf4ah5dpxeoozeysb24eke
« Previous Showing results 1 — 15 out of 25,791 results