Statistics & Computer Science
Permanent URI for this collectionhttp://repository.kln.ac.lk/handle/123456789/3751
Browse
8 results
Search Results
Item Information Technology education in the Sri Lankan school system: Challenges and perspectives(Sabaragamuwa University of Sri Lanka, 2006) de Silva, C.S.Item forecasting exchange rates using artificial neural networks(Journal of Applied Statistics, 2009) Chandrasekara, N.V.; tilakaratne, C.D.Item Human Perception of Haptic Force Direction(IROS 2006 IEEE/RSJ International Conference, 2006) Elhajj, I.; Weerasinghe, K.G.H.D.; Dika, A.; Hansen, R.In this paper we investigate the accuracy of human perception of haptic force direction applied to the hand. Haptic interfaces are commonly used in many applications and understanding the limitations of human perception would facilitate the design of these interfaces and the associated applications. The literature contains work related to force perception; however, none of which address the issue of the accuracy of haptic force direction perception. We discuss the design and implementation of the experiment used to evaluate the accuracy. Also presented are results related to training effects, fatigue and accuracy across angular regionsItem Analysis of key management in wireless sensor networks(IEEE EIT?07, 2007) Dustin, M.; Shankarappa, J.; Petrowski, M.; Weerasinghe, K.G.H.D.; Fu, H.A multitude of wireless sensor networks exist today in various fields, each having a specific objective in mind. Based on the objectives for each network, the security concerns can be different, dependent on such factors as the level of secrecy of the data being captured, the amount of computation done to the captured data, and the criticality of the data being available when needed. This paper aims to identify the various types of WSNs in existence today, review some of the key management schemes proposed by the community, and map each type of WSN to a set of these key management schemes that would be ideal to handle the security requirements for that network. Through our research, we aide in solving the question as to whether or not there exists any specific security concerns which are prevalent in a majority of WSNs in use today.Item Data centric adaptive in-network aggregation for wireless sensor networks(IEEE/ASME International Conference, 2007) Weerasinghe, K.G.H.D.; Elhajj, I.; Krsteva, A.; Najm, M.A.This paper presents and evaluates a data centric adaptive in-network aggregation algorithm for wireless sensor networks. In-Network data aggregation is used in wireless sensor networks to reduce the power consumption of sensor nodes. The accuracy of the aggregated results is highly sensitive to delays in the measurements. All existing methods use fixed time limit to accept delayed information for aggregation. The proposed method dynamically calculates the delay limit by using the historical behavior of each sensor. The presented simulation results illustrate the advantage of the developed algorithm.Item Preventing Cooperative Black Hole Attacks in Mobile Ad Hoc Networks: Simulation Implementation and Evaluation(International Journal of Software Engineering and Its Applications, 2008) Weerasinghe, K.G.H.D.; Fu, H.A black hole attack is a severe attack that can be easily employed against routing in mobile ad hoc networks. A black hole is a malicious node that falsely replies for any route requests without having active route to specified destination and drops all the receiving packets. If these malicious nodes work together as a group then the damage will be very serious. This type of attack is called cooperative black hole attack. In [9], we proposed a solution to identifying and preventing the cooperative black hole attack. Our solution discovers the secure route between source and destination by identifying and isolating cooperative black hole nodes. In this paper, via simulation, we evaluate the proposed solution and compare it with other existing solutions in terms of throughput, packet loss percentage, average end-toend delay and route request overhead. The experiments show that (1) the AODV greatly suffers from cooperative black holes in terms of throughput and packet losses, and (2) our solution proposed in [9] presents good performance in terms of better throughput rate and minimum packet loss percentage over other solutions, and (3) our solution proposed in [9] can accurately prevent the cooperative black hole attacks.Item A methodology for identifying ?Web metrics? for ?Web analytics?(2006) Wickramarachchi A P R; Malavisooriya L MThe need for perusing sound business practices when adopting an organizational web strategy is highlighted by many authors. Measuring effectiveness of business activities is a standard practice of any organization. However, few organizations actively take steps to assess successfulness of their web strategy. The main reason for this is due to the lack of a standard approach to measure the effectiveness of web sites. Many authors have come up with different types of metrics (measurements) for measuring the effectiveness of web sites. Two common approaches identified are; user centric measurements which depend on consumer surveys and site centric measurements which depend on server logs. The latter approach is preferred by many organizations Software tools have also been developed for collecting and analyzing results from web metrics. But still Organizations are faced with problems of identifying appropriate metrics for their web site. Different sections of a web site have different functions and objectives. Therefore different metrics are needed to measure the effectiveness of different sections of a web site. Thus a formal methodology for identifying suitable web metrics for a particular web site is invaluable for measuring the effectiveness of web sites. The research proposes a methodology for identifying appropriate web metrics for measuring the effectiveness of a web site; based on the objectives and functionalities of the web site. A set of metrics have been identified by combining the findings of previous researches which are evaluated through expert opinions. These metrics are the input for methodology development where a step wise approach is proposed to identify appropriate web metrics based on the functions of a particular web site.Item The Status of Business Continuity & Crisis Management (BCCM) in Large Scale Financial Information Systems in Sri Lanka(2007) Herrath, H.M.P.S.; Wijayanayake, W.M.J.I.The use of Information and Communication Technology (ICT) is on the increase and encompasses nearly the entire business process of Financial Industry. In this industry the rising competition and customer expectations have compelled management to implement and continuously upgrade scalable ICT practice and solutions. The range of financial services combined with the complexity of integrated ICI enabled delivery mechanisms require comprehensive partnership to be forged between Financial Organisations and ICT solution providers, especially with regards to Business Continuity & Crisis Management (BCCM). An effective BCCM model guarantees information availability and plays a vital role in ensuring an organization's survivability to run their business on 24*7. Such a model is critical as it will also provide reliable assurance on ensuring continuity of critical business functions in the event of crisis or disasters. Despite its importance, there has been no much research done to identify the current status of BCCM in Financial Organizations in Sri Lanka. This paper focuses on evaluating current status of BCCM in large scale financial information systems in Sri Lanka. Disaster tolerance, Disaster recovery planning, Digital systems security and Application Availability have been identified as the main four factors in BCCM. A questionnaire was developed and administered among total number of 100 executive levels and above staff members from financial industry to gather information related to the above four factors. The survey results indicate that 36�% of executive management aware or involves in BCCM programs. But only 6% of them has valid BCCM programs and rest of the organizations do not have acceptable BCCM program in place. 48% of the respondents stated that their organizations have no BCCM leadership and they are unable to answer the questions. Majority of managers (53%) believe Backup/Restore policy is the BCCM. One (1) out of Four ( 4) organizations has a Business Continuity Management (BCM) or BCCM in place against crisis or any kind of business functions discontinuity events. 17% of survey participants only knows their organisation has a BCCM and only 29% of managers have BCCM education in their organisation. Research reveals that a lot of Sri Lankan businesses, especially financial industry, is yet to consider systematically about BCCM and uncertainties in their environments.