ABSTRACTS OF RESEARCH PAPERS, 19TH FEBRUARY 2014 Develop a Detection System for Colour Stego Images through Extracting Colours-Gradient Cooccurrence Matrix Ahd Aljarf, Faculty of Engineering and Computing, Coventry University Recently, the concept of ‘Image Steganography’ is became an important issue in the computer security world. Image steganography simply means hide some secret data into an object. The object can be a text, an image or a sound, but the most popular cover object used for hidden secret message is images. A developed detection system is introduced in this research. The first part of the work includes creating varieties of stego-images. These stego-images are having different image file formats. Also, these stego-images have been done using different steganography methods and tools. The created stego-images are used to train the detection system in the next stage. On the other hand, the second part of the work includes detecting the hidden data. In order to do so the colour-gradient co-occurrence matrix (CGCM) is created for all tested images. The CGCM considers the information of both colour-correlation and gradient among the pixels in an image. Numbers of image features are extracted from the matrix. The system at this stage targeted the indexed images. The BMP images with 256 colours have been used to test the system. However, more image formats are planning to be used. Later, a classification method will be used to validate the features selected. These features are needed to differentiate between the clean images and the stego-images. An Ecosystem for E-learning in Mechatronics:The CLEM project Antonios Nanos, Faculty of Engineering and Computing, Coventry University This poster describes some results of the EU Leonardo Lifelong Learning Programme project CLEM (Cloud E-learning for Mechatronics). Following a European recognition for the need for updating of vocational education trainers in the area of mechatronics, CLEM set out to develop a cloud based teaching and learning facility for mechatronics based on the ideas of community, sharing and Web 2.0. An interesting development has been the creation of remote laboratories in the cloud. Learners can access such laboratories to help with their practical learning about mechatronics without need to set up laboratories at their own institutions. On the other hand the cloud infrastructure enables multiple laboratories to come together virtually to create an ecosystem for educators and learners. From such a system, educators can pick and mix materials to create suitable courses for their students. This means that learners can experience different types of devices and laboratories through the cloud. The paper provides an overview of the CLEM project, presenting results so far. In addition to the remote laboratories set up using a Raspberry Pi and Arduino microprocessor structure, this poster also presents information on the results of a survey and the development of static learning material. It also explains how a holistic e-learning experience can be obtained through use of static and dynamic material together with facilities for collaboration and innovation.. Annual Faculty Research Symposium - Abstract 1 GROUP-BASED PARALLEL MULTISCHEDULE FOR GRID COMPUTING Goodhead Tomvie Abraham, Faculty of Engineering and Computing, Coventry University As hardware design philosophy shifts from increasing the number of transistors and or increasing the clock speeds to the current multicore systems, shouldn’t there be a paradigm shift in software design philosophy? Multicore computers are already becoming pervasive in our daily computing environment while the scheduling challenges for Grid computing remains a challenge. Given that advances in computer hardware design (multicores) has automatically extended an open invitation to a shift in programming paradigm as it is a common knowledge that sequential programming does not scale well with multicores, neither do parallelism gain from sequential programming; should Grid scheduling be allowed to suffer the bottlenecks orchestrated by serial schedulers especially when the number of jobs increases? To avoid this problem hampering Grid scheduling and the growth of the Grid, it becomes imperative to exploit parallelism and multi-scheduling in Grid computing. Previous work in Grid scheduling has mostly addressed QoS, Makespan or the scheduling of parallel jobs and has failed to address the development and adoption of parallel schedulers that can take advantage of the multicores that is gradually taking over our computing lives. This work deals with the uses grouping methods to categorize both Grid jobs and Grid machines at Grid sites and employs the parallelism to multi-schedule Grid jobs and is aimed at exploiting parallelism - optimize the utility of several processors within a system, harness the advantages of the multicores and also scale Grid scheduling and sustain the growth of the Grid. Tests against the MinMin, MinMax and MinMinMinMax algorithms yielded far better scheduling times. And a Brief Introduction Computer hardware design philosophy has changed from increasing the number of transistors and clock speeds - which was good for sequential programming to multicores due largely to advances in circuit technology but programming model has not changed much to take advantage of this changes despite the established fact that sequential programming does not scale well with multicores. Grid computing entails that jobs are submitted by users in a location and processed at different locations millions of miles away. Grid computing is expanding at an extraordinary pace just as multicore systems are becoming ever more pervasive, the scheduling of Grid jobs will be hampered if not scaled in lieu with the two pronged innovation in hardware multicore computing and the expanding growth in the Grid. Multischeduling offers the ability to schedule multiple jobs from same source to various computing resources while Parallelism enhances the optimized use of all processing components within a system. To make good the revolution in computer hardware design, it has become imperative for a paradigm shift in software programming model A study of evaluation methodologies and impact of STEM Outreach activities Yamuna Bagiya, Faculty of Engineering and Computing, Coventry University The government of United Kingdom has identified the uptake of Science, Technology, Engineering and Mathematics (STEM) subjects into further education as a key factor in ensuring a successful future for the nation (HM Treasury 2004, The Work Foundation 2011). Also, research conducted by The Confederation of British Industry (CBI) has predicted that, in the next 5 to ten years, an additional 2.2 million people will be required in STEM related industry (CBI 2011). However, it has been indicated that there is a shortage in the supply of STEM graduates (Department of Education 2011, HM Treasury and BIS 2011, Roberts 2002) which will affect the economy negatively. To address the shortage of STEM graduates, and in order to increase the interest of younger generation in STEM subjects, many outreach and enrichment initiatives are being designed and Annual Faculty Research Symposium - Abstract 2 delivered. Some outreach and enrichment programmes are designed to give students the motivation and interest in science and mathematics from an early age in order to aid them to take up STEM related courses at university level (Toland 2011). They create awareness and encourage more students to take up STEM related courses (Department of Education 2011, Toland 2011). Younger students especially enjoy making exciting things, taking part in fun activities and experimenting with different things. The main purpose of developing outreach programmes is that the students can understand, relate and experience the fun behind STEM subjects (Jeffers et al 2004). However, so far there are no significant research findings on pedagogy of outreach programmes, methodology of longitudinal evaluation of different types of outreach activities and their impact. This paper will highlight the findings of research carried out at Coventry University, on evaluation strategies and the impact of STEM outreach activities on different widening participation groups. References 1.HM Treasury. (2004) Science & Innovation Investment Framework 2004 – 2014: Next steps. London: HM Treasury . 2.CBI. (2011) Building For Growth: Business priorities for education and skills. London: CBI 3.Roberts, G. (2002) SET for success: The supply of people with science, technology, engineering and mathematics skills: The report of Sir Gareth Roberts’ Review. London: HM Treasury 4.Toland, A. (2011) Employability Skills Review. Birmingham: The National HE STEM Programme 5.Department of Education. (2011) The STEM cohesion programme: final report. London: Department of Education 6.Jeffers, A T., Safferman, A G., and Saffermsn S I. (2004) ‘Understanding K-12 Engineering Outreach Programs’. Journal of professional issues in engineering education and practice 130, 95-10 Network Intrusion Detection Systems (NIDS) for Multi Attack Scenarios in Multi-Core Environments in Computer Networks Waleed Bul’ajoul, Faculty of Engineering and Computing, Coventry University With growth of various malicious attacks, especially technology is more dependent on network and wireless systems, the traditional security tools such as anti-virus and firewalls are not sufficient to provide the integral, reliable and security free networks. Due to the high and precise detection from the malicious attacker, intrusion detection is one of the strong technologies that monitors network traffic and recognize illegal use, misuse of computer system, and malicious attackers. This study outlines a NIDSs software exploit that utilizes QoS and parallel NIDSs technologies in Cisco Catalyst Switches to increase NIDSs analytical performance when deployed in high-speed networks. We have designed a real network to present experiments that use a Snort NIDS to demonstrate the weaknesses of NIDSs, such as its inability to process multiple packets and its propensity to drop packets in heavy traffic and high-speed networks without analysing them. We tested Snort’s performance analysis, gauging the number of packets sent, analysed, dropped, filtered, injected, and outstanding. We suggest using QoS configuration technologies in a Cisco Catalyst 3560 Series Switch and parallel Snort NIDSs to improve NIDS performance and to reduce the number of dropped packets. The proposed research will design and implement a Network Intrusion Detection System(NIDS) to improve network security in terms of efficiency , effectiveness, performance analysis and detection engine. Annual Faculty Research Symposium - Abstract 3 Implementing median filter in 2D medical images using FPGA architecture Razan Aboabat, Faculty of Engineering and Computing, Coventry University FPGA becomes commonly used in biomedical application in different fields such as, brain tumor detection with MRI images whereas its strength comes from the flexibility that derives from the programmable environment which is the main difference with SIMD. In the median filter operation, FPGA with pipeline technique is efficient compared with CNAPS and IMAP-Vision using specific algorithm to optimize the performance by reducing the comparison times from 30 to 19 in order to achieve efficiency almost by 33%. The main focus of the research is to solve large-scale image processing problems with a fast response time with data size of 100 images size of (1024*1024) using PVM message passing system environment considering synchronization issue between different processes belonging to the same or different programs in parallel computing. Educational Game: Giving Tutorial and Exercise for Network Subnetting Mohammad Adnan Hijji, Faculty of Engineering and Computing, Coventry University Subnetting is an integral part of the network computing technique, and a very significant process for the network of any organization, where it plans and arranges the company’s computer network. In fact, students who studied networking need to practice this concept during and after networking course, in order to gain and understand this concept effectively. Attracting students to practice Subnetting technique needs modern and entertaining methods of giving tutorials and exercises. From this perspective, the idea of this project came up. Hence, this research presents an educational game which aims to support lecturers in computer networking classes. The proposed game provides tutorials and exercises regarding the concept of ‘Subnetting technique’ in an entertaining manner. The design of game takes into account the targeted audience, their pre-required knowledge and Human Computer Interaction (HCI). The Storyboard’s method was used to design the interfaces sequence of the game, while Unified Modelling Language (UML) was used to design the software. The designed framework of the game was implemented by C# programming language. The game was constructed, tested and evaluated successfully. The evaluation process was through a process of primary data collection. Keywords- Educational Game; Subnetting Game; Converting Decimal to Binary Game; Converting Binary to Decimal Game; IP address classes Game; IP hosts Game Annual Faculty Research Symposium - Abstract 4 Investigation of the current E-Learning Situation in Higher Education in Saudi Arabia Latifah Almuquren, Faculty of Engineering and Computing, Coventry University The use of IT/ICT in the field of education has become remarkable across many countries. The unification between these fields (i.e. IT/ICT and education) has enhanced education across several areas, such income and teaching methods. Public and private organizations are required on-going development and evaluation in order to find out the most effective approach of education across various levels of education. In the sight of government's interest in Saudi Arabia, toward the development in higher education including E-learning, this research aim to investigates the current status of E-learning in universities within Saudi Arabia through exploring barriers and hurdles that facing deployment E-learning in universities within Saudi Arabian, undertaken the experiences of the UK and Australia in E-learning to propose a set of recommendations for the sake of developing Elearning in Saudi Arabian universities. The results indicate that there are (i) lack of applications Elearning in most of universities (ii) and they missed some of the major requirements of E-learning which aid to configure and implement E-learning, although it has been implemented for more than five years. Active Buckling Control for Vehicle Body Structures James E Trollope, Control Theory and Applications Centre (CTAC), Coventry University An active buckling control strategy for mechanical structural applications is being developed at the Control Theory and Applications Centre (CTAC), Coventry University, whereby the structures are designed to exhibit a natural response phenomenon when subject to external stimuli, e.g. an applied force. An example of a natural structure is the self-preserving properties of trees where they are able to automatically flexibly change their shapes in response to dynamic loads, e.g. to counteract effects of wind and adapt to the variation in humidity levels. Applying such an approach to mechanical structures provides an opportunity to reduce the mass of current structures, which are typically over-engineered in order to deal with undesirable crash scenarios. This poster presents the concept of a novel active buckling control strategy aimed at alleviating the compatibility problem arising when two vehicles of dissimilar mass and stiffness values encounter a collision. This approach assumes that the properties of the vehicle body structures can be changed via actively controlled materials, commonly known as smart materials; in particular the smallest positive eigenvalue which corresponds to the point at which buckling commences. This approach is based on a multi-dimensional look-up table combined with fuzzy logic for interpolating between precalculated levels of energy absorption related to the stiffness values. By modelling the force versus deformation characteristic, the energy to be absorbed is more appropriately apportioned, thus enhancing vehicle safety. Annual Faculty Research Symposium - Abstract 5 Dynamic Resource Allocation for Web Hosting Environment Olfat Mirza, University of Warwick Hardware and system resource allocation to e-business web application has been a known challenge and it is becoming increasingly difficult to allocate resource efficiently beforehand as there is variation in workload received by these applications that is difficult to predict. Ignoring this fact can either lead to over-allocating expensive resources inefficiently, or it can lead to have long response time that anticipated, frequent request timeouts, or unexpected connection closure or system failure. To mitigate this risk, resources can be re-allocated dynamically bearing in mind the the benefit or resource re-allocation exceed the cost of the re-allocation. One possible way to achieve this is to host these applications in virtualised server pools, and dynamically reassign computer resources between pools to meet the demands on the hosted application. In this work, a combination of the reactive behaviour of two well-known switching policies (Proportional Switching Policy and Bottleneck Aware Switching Policy) is proposed with the proactive forecasting of expected workload using seven forecasting models. These forecast models are to be implemented and test via simulation to gauge the effectiveness achieved. Annual Faculty Research Symposium - Abstract 6