The Experts below are selected from a list of 93 Experts worldwide ranked by ideXlab platform
Jens Haag - One of the best experts on this subject based on the ideXlab platform.
-
DVCL : A Distributed Virtual Computer Lab for Security and Network Education
2018Co-Authors: Jens HaagAbstract:Teaching Networking and IT security in higher education requires a safeplayground for students, where they can safely carry out hands-on exercises.This safe playground is known as a computer lab. Universities have todesign and to provide such a lab with respect to certain criteria, e.g.technical opportunities, educational requirements and demands of thelearners. Since there is no one-size-fits-all lab, the labs will be designed tofit into a certain context and thus have own strengths and weaknesses.In this thesis, we investigate and work with two established labs, whichwere designed for hands-on experiences in Networking and it security courses.These labs are predominantly different but have an essential overlap ineducational requirements.One lab is developed by the Open University. It is dedicated for distancelearning. It is based on virtualization and every student is able to startthis lab on his own computer. Students can work out exercises wheneverand wherever they want. A shortcoming however is that students have towork alone, (distant) group work is not possible due to the isolated labarchitecture. This lab is the technical base for our research.The other lab is developed by the Cologne University. It is a physicallab, dedicated for on-campus courses and thus it is not portable. Butstudents can meet in the lab, work in groups and are able to get supportfrom a course advisor, who is also able to verify exercises. A shortcominghowever is that students must be present (they have to travel to theuniversity) and they are dependent on the opening hours of the laboratoryand the availability of the course advisers.In two research parts, we show how such two different lab approachescan be combined and what can be achieved.The first research part is about design issues.Initially, we enable group work in our lab for distance education, sincegroup work is an essential part in on-campus classes. Also remote studentsshould be able to work together. Since the lab is designed as an isolatedsystem, the challenge is to connect two of them on the Network level butwithout creating a potential bridge between the isolated and the outsideworld. We achieved this by adding a communication interface to the labarchitecture. This communication interface consists of a ghost host toextract and inject Network packets, and a remote bridge endpoint, totransport these packets between remote ghost hosts across an intermediateconnection, e.g. the internet. The developed prototype is called DistributedVirtual Computer Lab (DVCL) and enables to connect two or more distantlabs while preserving the isolated character.The DVCL is then extended and improved by a central authority (CA).While the point-to-point connection of the communication interface canconnect two remote Networks in a handy way, more connections requirecareful planning by the students. We show that a CA simplifies the usageof our DVCL for the students (and also for academic staff) and in additionto it avoids administrative configuration errors while connecting remotelabs, e.g. a circular flow which leads to an unusable lab.The first part is completed by two applicability enhancements. Thefirst enhancement covers and resolves security issues in order to pushour prototypical implementation of the DVCL and the CA closer to aproductive learning environment. The second enhancement introduces aGraphical User Interface to increase the usability of the DVCL.The second research part is about educational aspects.In the first part, we assume that working independent from a physicalon-campus lab as well as group work is essential for our students. Ourevaluation of more than 200 students participating in an on-campus Networkingcourse shows, that nearly half of the students actually say, thatthey would like to work independently from the university at least partiallyand they would welcome the introduction of an e-learning system. Inaddition, a predominant majority think of working in groups as well asreceiving guidance and feedback as crucial to their learning success. Thisresult justifies and confirms our research and also reveals an additional requirement.The challenge is to provide feedback and guidance to a student, who isworking on an exercise and a human course advisor is not available. Thisis e.g. when students use the DVCL at home in the evening hours. Weshow, that Captured Network Traffic of a lab can give some indication ofwhat a student has already configured according to a certain exercise. Weuse this insight to develop an Electronic Exercise Assistant. This softwareprogram is able to recognize the progress of an exercise and can provideappropriate feedback and support, based on preloaded rules and conditions.This significantly improves the learning situation for students workingremotely in lab. Besides this automatic support, the exercise assistant canverify intermediate and complete solutions of an exercise.The second part is completed by an educational enhancement. Ourevaluation and also own observations show, that a lab is more than aroom with computer and Network facilities. Rather it is a social placewhere students e.g. meet, form learning groups, talk and discuss. Weuse these insights and enhance the DVCL to support social interactions.Based on our on-campus lab as source, we model a set of communicational,organizational as well as educational activities and implement them in ourDVCL. The result shows, that our DVCL prototype is no longer a technicalplatform but a virtual place, where students can meet, communicate,arrange learning groups, exchange experiences and work on exercises.This thesis shows that aspects of our two different lab environments canbe combined. Our resulting Distributed Virtual Computer Lab incorporatesstrengths of each source lab. It is a gain for distance teaching as well as foron-campus classes. Remote students are now able to utilize the lab being avirtual classroom, where they can learn in groups, assisted by an electronicadvisor and without the need for a face-to-face meeting. On-campus classescan offer students a new learning environment, where they can learn in aclassroom character without the need to travel to the university.
-
DVCL : A Distributed Virtual Computer Lab for Security and Network Education
2018Co-Authors: Jens HaagAbstract:Teaching Networking and IT security in higher education requires a safeplayground for students, where they can safely carry out hands-on exercises.This safe playground is known as a computer lab. Universities have todesign and to provide such a lab with respect to certain criteria, e.g.technical opportunities, educational requirements and demands of thelearners. Since there is no one-size-fits-all lab, the labs will be designed tofit into a certain context and thus have own strengths and weaknesses.In this thesis, we investigate and work with two established labs, whichwere designed for hands-on experiences in Networking and it security courses.These labs are predominantly different but have an essential overlap ineducational requirements.One lab is developed by the Open University. It is dedicated for distancelearning. It is based on virtualization and every student is able to startthis lab on his own computer. Students can work out exercises wheneverand wherever they want. A shortcoming however is that students have towork alone, (distant) group work is not possible due to the isolated labarchitecture. This lab is the technical base for our research.The other lab is developed by the Cologne University. It is a physicallab, dedicated for on-campus courses and thus it is not portable. Butstudents can meet in the lab, work in groups and are able to get supportfrom a course advisor, who is also able to verify exercises. A shortcominghowever is that students must be present (they have to travel to theuniversity) and they are dependent on the opening hours of the laboratoryand the availability of the course advisers.In two research parts, we show how such two different lab approachescan be combined and what can be achieved.The first research part is about design issues.Initially, we enable group work in our lab for distance education, sincegroup work is an essential part in on-campus classes. Also remote studentsshould be able to work together. Since the lab is designed as an isolatedsystem, the challenge is to connect two of them on the Network level butwithout creating a potential bridge between the isolated and the outsideworld. We achieved this by adding a communication interface to the labarchitecture. This communication interface consists of a ghost host toextract and inject Network packets, and a remote bridge endpoint, totransport these packets between remote ghost hosts across an intermediateconnection, e.g. the internet. The developed prototype is called DistributedVirtual Computer Lab (DVCL) and enables to connect two or more distantlabs while preserving the isolated character.The DVCL is then extended and improved by a central authority (CA).While the point-to-point connection of the communication interface canconnect two remote Networks in a handy way, more connections requirecareful planning by the students. We show that a CA simplifies the usageof our DVCL for the students (and also for academic staff) and in additionto it avoids administrative configuration errors while connecting remotelabs, e.g. a circular flow which leads to an unusable lab.The first part is completed by two applicability enhancements. Thefirst enhancement covers and resolves security issues in order to pushour prototypical implementation of the DVCL and the CA closer to aproductive learning environment. The second enhancement introduces aGraphical User Interface to increase the usability of the DVCL.The second research part is about educational aspects.In the first part, we assume that working independent from a physicalon-campus lab as well as group work is essential for our students. Ourevaluation of more than 200 students participating in an on-campus Networkingcourse shows, that nearly half of the students actually say, thatthey would like to work independently from the university at least partiallyand they would welcome the introduction of an e-learning system. Inaddition, a predominant majority think of working in groups as well asreceiving guidance and feedback as crucial to their learning success. Thisresult justifies and confirms our research and also reveals an additional requirement.The challenge is to provide feedback and guidance to a student, who isworking on an exercise and a human course advisor is not available. Thisis e.g. when students use the DVCL at home in the evening hours. Weshow, that Captured Network Traffic of a lab can give some indication ofwhat a student has already configured according to a certain exercise. Weuse this insight to develop an Electronic Exercise Assistant. This softwareprogram is able to recognize the progress of an exercise and can provideappropriate feedback and support, based on preloaded rules and conditions.This significantly improves the learning situation for students workingremotely in lab. Besides this automatic support, the exercise assistant canverify intermediate and complete solutions of an exercise.The second part is completed by an educational enhancement. Ourevaluation and also own observations show, that a lab is more than aroom with computer and Network facilities. Rather it is a social placewhere students e.g. meet, form learning groups, talk and discuss. Weuse these insights and enhance the DVCL to support social interactions.Based on our on-campus lab as source, we model a set of communicational,organizational as well as educational activities and implement them in ourDVCL. The result shows, that our DVCL prototype is no longer a technicalplatform but a virtual place, where students can meet, communicate,arrange learning groups, exchange experiences and work on exercises.This thesis shows that aspects of our two different lab environments canbe combined. Our resulting Distributed Virtual Computer Lab incorporatesstrengths of each source lab. It is a gain for distance teaching as well as foron-campus classes. Remote students are now able to utilize the lab being avirtual classroom, where they can learn in groups, assisted by an electronicadvisor and without the need for a face-to-face meeting. On-campus classescan offer students a new learning environment, where they can learn in aclassroom character without the need to travel to the university.
S. Martin - One of the best experts on this subject based on the ideXlab platform.
-
High precision Traffic measurement
IEEE Communications Magazine, 2002Co-Authors: John G. Cleary, Ian Graham, Tony Mcgregor, Murray W. Pearson, L. Ziedins, J. Curtis, Stephen F. Donnelly, Jed Martens, S. MartinAbstract:Making passive measurements is a challenging process. A small number of research groups have overcome these difficulties and have made terabytes of Captured Network Traffic available to the Internet community. We have discovered that even when initial hurdles to collecting traces have been overcome, there are potentially many methodological weaknesses that can severely limit the usefulness of the data collected. This article describes these weaknesses and suggests methodologies to avoid them. Five case studies from the work of the University of Waikato's WAND Network research group are presented. These studies illustrate the need for care at all stages of the measurement process, analysis, and presenting the results.
Sanmeet Kaur - One of the best experts on this subject based on the ideXlab platform.
-
Extreme Gradient Boosting Based Tuning for Classification in Intrusion Detection Systems
Communications in Computer and Information Science, 2018Co-Authors: Ashu Bansal, Sanmeet KaurAbstract:In a fast-growing digital era, the increase in devices connected to internet have raised many security issues. For providing security, varieties of the system are available in the IT sector, Intrusion Detection system is one of such system. The design of an efficient intrusion detection system is an open problem to the research community. In this paper, various machine learning algorithms have been used for detecting different types of Denial-of-Service attack. The performance of the models have been measured on the basis of binary and multi-classification. Furthermore, parameter tuning algorithm has been discussed. On the basis of performance parameters, XGBoost performs efficiently and in robust manner to find an intrusion. The proposed method i.e. XGBoost has been compared with other classifiers like AdaBoost, Naive Bayes, Multi-layer perceptron (MLP) and K-Nearest Neighbour (KNN) on recently Captured Network Traffic by Canadian Institute of Cybersecurity (CIC). In this research, average class error and overall error have been calculated for the multi-classification problem.
Žarko Čučej - One of the best experts on this subject based on the ideXlab platform.
-
Modeling of Captured Network Traffic by the mimic defragmentation process
SIMULATION, 2010Co-Authors: Matjaž Fras, J. Mohorko, Žarko ČučejAbstract:Over recent years, in order to assist the evaluation, construction, and upgrade of communication Networks, the need for the simulation of complex communication Networks has increased. Traffic modeling has had a very large impact on Network simulation reliability, which is usually statistically described for simulation purposes. Most often, the Network Traffic analysis is based on the Captured packets. However, modeling of Network Traffic is usually described by statistics of data sources, from higher layers of a Transmission Control Protocol/Internet Protocol (TCP/IP) model. For these reasons, we have developed a method that allows distribution parameter estimation of the data-source process from Captured packets. This method is based on the algorithm, which mimics defragmentation as opposite to the TCP/IP fragmentation and encapsulation processes. The proposed method achieves an accurate description of Network Traffic.
Mustapha Belouch - One of the best experts on this subject based on the ideXlab platform.
-
Distributed intrusion detection system for cloud environments based on data mining techniques
Procedia Computer Science, 2018Co-Authors: Mohamed Idhammad, Karim Afdel, Mustapha BelouchAbstract:Nearly two decades after its emergence, the Cloud Computing remains gaining traction among organizations and individual users. Many security issues arise with the transition to this computing paradigm including intrusions detection. Intrusion and attack tools have become more sophisticated defeating traditional Intrusion Detection Systems (IDS) by large amount of Network Traffic data and dynamic behaviors. The existing Cloud IDSs suffer form low detection accuracy, high false positive rate and high running time. In this paper we present a distributed Machine Learning based intrusion detection system for Cloud environments. The proposed system is designed to be inserted in the Cloud side by side with the edge Network components of the Cloud provider. This allows to intercept incoming Network Traffic to the edge Network routers of the physical layer. A time-based sliding window algorithm is used to preprocess the Captured Network Traffic on each Cloud router and pass it to an anomaly detection module using Naive Bayes classifier. A set of commodity server nodes based on Hadoop and MapReduce are available for each anomaly detection module to use when the Network congestion increases. For each time window, the anomaly Network Traffic data on each router side are synchronized to a central storage server. Next, an ensemble learning classifiers based on the Random Forest is used to perform a final multi-class classification step in order to detect the type of each attack. Various experiment are performed in the Google Cloud Platform in order to assess the proposed system using the CIDDS-001 public dataset. The obtained results are satisfactory when compared to a standard Random Forest classifier. The system achieved an average accuracy of 97%, an average false positive rate of 0.21% and an average running time of 6.23s.