Search

The growth in data traffic is exponential with the usage of network applications in mundane activities. The network operators are posed with the challenge of providing Quality of Experience (QoE) to the deluge of internet users. Network traffic classification plays a significant role in network resource management with prominent security, billing, and accounting applications. In this paper, the network data traffic classification is performed using the Deep Learning (DL) models. The previous data traffic mechanism fails when the scale in data generated is in Exabytes per day from the perspective of the internet user. The network resource management is automated by classification of the traffic without intervening by the operator. The dataset is collected from the campus network for different applications.The network traffic classification is performed using the AlexNet, ResNet, and GoogLeNet DL models. The accuracy obtained for ResNet is 95%, AlexNet is 75%, and GoogLeNet is 91%. The challenge in network traffic classification is converting the packet capture files to the data type requirements of the Convolution Neural Networks (CNN) as an image. The four different network applications are considered for traffic classification. In the next-generation network architecture, artificial intelligence will be an integral part. It reduces the human intervention in traffic characterization and analysis, which aids in meeting the QoE requirements of the users.

Various IP crawlers are available using which an attacker can get IPs of vulnerable edge devices, gain access, manipulate and misuse the devices causing mild to potentially threatening events in the IoT network. As IoT technology has spread its wing in different sectors such as manufacturing, transportation, healthcare, agriculture and industry, securing these devices is crucial. The automation of operations is performed using robots in the evolving sector Industrial IoT generating a high need for secure communication. The attacker can scan for vulnerable edge devices using the IP crawlers available on the internet and access these devices. The MQTT protocol is used for data transfer between things and brokers. The log analysis of the "Message Queue Telemetry Transport" (MQTT) protocol is performed by generating the "connect," "publish" and "subscribe" logs. These logs provide information related to the protocol name, topic names and Quality of Service (QoS) levels used for communication. The inference derived from the MQTT protocol log analysis clearly shows that the attacker has gained access to the edge device. Then the attacker can subscribe as well as publish on the topic names obtained. When the attacker can publish on the vulnerable edge device's topic names, it exposes the IoT network for threat events. The objective is to prevent exposure of the edge devices from threat events leading to mismanagement of commands given to the IoT devices. We present the measures to prevent the threats by MQTT protocol log analysis.

With the explosive growth of IoT applications, billions of things are now connected via edge devices and a colossal volume of data is sent over the internet. Providing security to the user data becomes crucial. The rise in zero-day attacks are a challenge in IoT scenarios. With the large scale of IoT application detection and mitigation of such attacks by the network administrators is cumbersome. The edge device Raspberry pi is remotely logged using Secure Shell (SSH) protocol in 90% of the IoT applications. The case study of SSH brute force attack on the edge device Raspberry pi is demonstrated with experimentation in the IoT networking scenario using Intrusion Detection System (IDS). The IP crawlers available on the internet are used by the attacker to obtain the IP address of the edge device. The proposed system continuously monitors traffic, analysis the log of attack patterns, detects and mitigates SSH brute attack. An attack hijacks and wastes the system resources depriving the authorized users of the resources. With the proposed IDS, we observe 25% CPU conservation, 40% power conservation and 10% memory conservation in resource utilization, as the IDS, mitigates the attack and releases the resources blocked by the attacker.

Networking Industry has wide opportunities for Computer Science graduates with the skills of cloud architecture, cyber security, Internet of Things, which are converging areas. Computer Network is a fundamental course in the Computer Science stream, has a challenge in teaching / learning the concepts of networking, which is considered to be a black box. The objective is to open up this black box content through Project-Based Learning (PBL) to better the conceptual understanding and enhance the 21st-century skills required to develop network applications and the services.The PBL activity has helped the students in contributing to developing real time network based solutions collaboratively.The analysis of pre and post-test data shows a 20% improvement in network concepts learned. The peer and self-assessment results demonstrate a 10% improvement in 21st century skills such as collaborative, critical thinking and creativity.

The Stock Market prophets focus on developing a efficient approach to predict the future stock prices. Apart from achieving best results in the successful stock market prediction it is also necessary to minimize the inaccurate forecast of stock prices. The project aims to design and implement a predictive system for guiding stock market investiture. As stock market is a very volatile environment, it can be influenced by positive or negative stock price releases. The research shows that there is an effect of prices on stock market and that the stocks can be predicted with the help of those prices. Automated data in general and company’s distinct data of Apple, Microsoft, Google and Amazon are used to test the effect of the efficient prices on the stock market. A different procedure is experimented with sentiment analysis and the results display that it is able to predict the stock market with the help of stock prices for different companies, where the linear regression model is possible to predict more accurately than the sentiment analysis method.

The network of computers infected with malicious software without the knowledge of owner is termed as Botnets. Cyber criminals use these botnets for malicious activities such as Distributed Denial of Service attacks, blagging of sensitive data. Recent botnets namely zeus, kraken, Nargache and many others apply a technique called DGA (Domain Generated Algorithms). DGA dynamically generates large number of random domains and choose subset for use so that static domains remains ineffective. The legitimate and DGA domains identification are often understood as most difficult to identify. The proposed work deals with the identification of legitimate domains from DGA generated ones using Classification models and N-Gram models. The introduction of N-Gram model with classification model achieves much accuracy.

Cloud computing technology usage has emerged as a prominent field in providing the computation services to the users using the internet. The 3As-AnyTime, AnyWhere and AnyDevice concept changed the industry towards user-centric needs to collaborate and connect with people, which makes software industries to adopt the cloud-based development environment. This created the scope for educationist to enable the students with the skills related to cloud computing technology. A course on distributed and cloud computing is introduced at the undergraduate level for catering to this. The authors discuss the challenges faced in teaching-learning cloud computing technology and the method followed to overcome these. The two main challenges are designing the course content and adopting suitable open source infrastructure tools for hands-on experience. The authors have put efforts in adoption and usage of cloud computing in undergraduate engineering level by proposing a modified Technology Acceptance Model (TAM). This model helped to identify different parameters/attributes to improve academic performance. An activity-based laboratory course is designed in line with the theory course. All the activities are implemented using the open source software environment. The activities planned in the course are assessed based on the outcome based evaluation. The graduate attribute addressed in the course are domain knowledge gain in cloud computing concepts, problem-solving, usage of modern tools and life-long learning. The average attainment of the stated graduate attribute is 77.89% in the course internal semester assessment. The effectiveness of proposed method is validated by defining hypotheses relating to the research questions framed. Results prominently speak about the advantage of TAM and its implementation in the course.

The emergence of the disruptive technologies such as cloud computing, IoT (Internet of Things), machine learning and data analytics has made its mark in different sectors like transportation, agriculture, healthcare, environment monitoring, renewable energy systems, retail, and industry. The IoT provides connectivity of things dynamically to the network whereas cloud computing provides virtualization in storage and processing. In CloudIoT paradigm, cloud and IoT are merged together to provide complementary features in smart applications/services. CloudIoT solutions make it possible to envisage ubiquitous and pervasive connectivity to the users. The source of livelihood for a majority of the population is agriculture; it mantles a dominant role in the economy of the country. The proposed work focuses on CloudIoT architecture for providing any smart solutions in different sectors. A case study of a smart irrigation system is discussed in the paper. A smart irrigation system is developed using the lightweight protocol, MQTT (Message Queue Telemetry Transport). MQTT protocol is 22% more energy efficient and 15% faster when compared with other protocols. The temperature and soil moisture data are collected and managed by Amazon cloud. The data analysis is performed using the Weka (Waikato Environment for Knowledge Analysis) tool. The cost-effective solution is demonstrated and results speak the strength and performance of the system.

Knowledge of Internet of Things (IoT) provides an opportunity to explore career possibilities, to gain experience in aspects of engineering beyond the core curriculum and it brings interdisciplinary mode of learning in students. This paper discusses the emergence of IoT in the curriculum, towards building human resources that contribute to the current issues related to smart cities. IoT is the network of all kinds of things embedded with sensors, electronics, software and people connected to the Internet. Gartner predicts that 6.4 billion connected things would be in use by the end of the year 2018 also it would reach 20.8 billion by 2020. The emergence of the IoT has a transformative direct effect on our society and forces us to rethink how to educate the next generation engineers. As a result universities, institutes have taken up a step to include loT as a course in the curriculum. The authors use the experience of teaching loT as an elective course and propose a framework for inclusion of loT as a course in the undergraduate curriculum. The outline of the course content, course outcomes, and delivery method are discussed in this paper. The graduate attributes 3: design/development of solutions, 5: modern tool usage, 6: Engineer and Society, 12: lifelong learning and 13: The program specific outcomes on domain knowledge of network engineering specified by National Board of Accreditation (NBA) are focused through this course.

The major challenge of Team Based Learning (TBL) or Group learning is its Design and Implement that assess the student's knowledge and skills. The challenge also includes designing assessment to grade the students based on their contribution to the team performance. Teacher assessing the student's performance in team is a teacher biased process. Therefore Peer evaluation can be included as part of TBL assessment. Peer review is considered as transparent and efficient technique to assess individual in team. But it has also been criticized with respect to fairness. This paper proposes Trusted Relative Peer Review (TRPR) which reduces the disadvantages of peer review to greater extent. It includes two reviews i) Team's Guide review ii) Peer review. The trustworthiness of peer review is decided based correlation of Guides and Peer reviews, depending on which each student is rated relatively for 100 points. After comparisons of results, 65% student teams peer reviews were found to be not trustworthy, which are corrected through proposed TRPR. This review is accepted by 54-58% of students and trusted it. TRPR helped to measure "ability to function in a team" with attainment of 7.7 on the scale of 10. TRPR contributes to the success of TBL implementation in greater extent by overcoming the drawbacks of peer review. It influences individual student's marks based on their contributions. Therefore it ensures students contribute in group work resulting in students learning. Although TRPR is hectic, future work shall try to make it simple.

The technological rebellion has made current generations to depend more on the digital world in their day to day life. The effectiveness of the digital world has changed the learner's perceptive towards the education system. The impulsivegrowth of technology has increased the popularity of Information and Communication Technology which has created a research scope in Computer network domain thus Computer network course in the curriculum of Computer Science and Engineering plays an important role. It is always a challenge to teach Computer Networks course to students as the course requires a practical exposure more than the chalk and talk. It is very hard for students to envision the concepts of computer networks, protocols formats and understanding packets flow during teaching in classroom. To overcome above said challenge, pedagogical practice such as active learning through simulation tool is identified in delivery of the course. This paper discusses the effective use of a Packet Tracer (PT) simulation tool in the teaching of computer network concepts. A structured learning approach is introduced by using packet tracer tool in teaching network concepts and series of exercises are designed and developed to give the visualization of each layer functionality. These practical exercises helped in active learning to use the packet tracer to simulate campus network. This pedagogy practice encouraged the students' to build the clarity on the physical devices like router, access points, switch and their configurations. The troubleshooting skills were practiced by introducing the errors deliberately and how to resolve is learnt through the tool. Computer network simulation activity include the concepts of NAT, Subnetting, V-LAN setting, services of TCP/UDP, packet format details, application layer protocol like DNS, DHCP, E-Mail, FTP and HTTP. Activity is evaluated through demos, presentation and quizzes. Rubrics based assessment is followed to ensure uniformity and transparencyin the evaluation. The average marks obtained in quiz by students before using PT tool is 65% and after using tool is 85%, it shows significant improvement of 20% in learning the fundamental concepts of computer network course.

The first challenge in offering this course is identifying the content of the course, at what level the course needs to be introduced, since both hardware and software components are involved in IoT. Second challenge is involving students in learning process as it is not only theory based and different knowledge level students are part of this course. To address these challenges we followed activity based approach for teaching and learning this course. Various real-world problems are identified, and assigned to different teams consisting of four students. Outcome based evaluation is followed to assess the group activity. The author's observations indicate that the activity based learning addresses graduate attributes of UG program, such as Design/Development of solutions, Modern tool usage, Engineer and society and Life-long learning. The average attainment of the mentioned graduate attributes is 79.18% for the course in continuous internal assessment.

User experience design is all about making a product by understanding its users and enables user to use a product easily. UX design is part of current industry standards which can be started along with requirement phase. Undergraduate students build products in project courses. Such courses include Software Life Cycle phases like Requirement gathering/ analysis, Design, Implementation and Testing but not UX design. Thus best of the project fail to generate a product that is usable. This paper presents a case study of minor project course which incorporated UX design in building mobile apps. UX is new technology and little known to instructors at colleges. So students were instructed to learn UX from MOOC on 'UX Design for Mobile Developer's', hosted by Udacity. Also, a workshop by industry expert in UX was conducted. The UX activities specified by MOOC were included as part course assessments. Through the UX design, students identified 3 to 4 user/ mobile device requirements and incorporated them in the implementation. The weightage of marks for UX design is 24%, which increased Design phase weightage to 30%. Finally students were able to launch product, 'KLE Tech STUDio' mobile-app. Attainment of design phase, UX, activities specified by MOOC, implementations are 7.2, 7.82, 6.8 and 7.5 respectively on the scale of 10. Further UX design can be included as part of Software Engineering course. Also, UX design can be tailored to any project course depending on the product being developed.

Massive Open Online Course (MOOC) is the best means to learn latest /current technologies which are otherwise difficult to gain access and learn them. A case study of how, MOOC on 'User Experience (UX) provided by Udacity for mobile app' was utilized to enhance the quality of student projects, is presented in this paper. Most of the time, MOOCs are provided by the global experts who teach the course in a best possible way. Also, MOOC directly helps in attaining lifelong learning skills. MOOC can be used in many ways to deliver a curriculum. This paper focuses on, how the learning's and activities specified by MOOC on UX is included as part of student project course, its assessment, evaluation and results. The student projects are mobile apps specified by a company, which emphasized much on UX, the latest industry need. Therefore students were instructed to register for the MOOC on UX and learn on their own. Each student teams had to do UX design as specified in MOOC and same UX design activities were included as a part of project course assessment and evaluated. As a result, students were able to discover additional requirements addressing mobile UX and perform UX based design and construct according to it. These activities helped to attain ABET criteria 'i' with attainment 7.82 on scale of 10.

Team-based learning (TBL) is a form of student-centered active learning which will help to develop various professional skills, problem solving, interpersonal team skills, and self-directed independent learning. The success of TBL approach depends on its design comprising Team formation, Project identification and distribution, monitoring the progress, maintaining open discussion in every phase and peer evaluation. This paper presents the case study TBL design done for minor project course. First step, the team formation plays a very important role in projects. Therefore the teams are formed based on the complementary skills of the student and ensuring that each team has an expert member in every identified skill. The next step is project distribution, while distributing the project the team strength is calculated and accordingly the problem statements are assigned. Also to increase the motivation and learning in the team, open discussions are held where each team can review and critic other teams work in every phase. The last part in the TBL is assessment and evaluation, which is incomplete without including peer evaluation. Although the Team work is evaluated by the instructor, the individual evaluation is done by both instructor as well as peer students within the team. This TBL design has improved the team's performance and each individual is able to work better in a team. Here students could successfully build and lunch mobile app as a product. The paper gives results of various TBL components. It concludes with challenges faced in this case study for further research.