Kalum Priyanath Udagepola

The sensors connectivity within SCADA automation environment and new trends for security development during multicasting routing transmission

This study examined the security of SCADA system and its protocols, more specifically, SCADA/DNP3 protocol security. To achieve the study goals, a SCADA simulation environment is designed for water pumping process through connectivity of intelligent sensors, the payload is constructed, and security is deployed inside DNP3 protocol stack and then bytes are multicast to subcontrollers.

Access-Pattern Aware Checkpointing Data Storage Scheme for Mobile Computing Environment

Abstract The rapid development of communication technology and mobile devices has facilitated the mobile computing applications such as location-based services, and information sharing. Characteristics of the mobile computing system, such as the space limitation, mobility handling, low bandwidth and limited battery life, make mobile computing applications more prone to failures. Checkpoint and rollback recovery technology, as a fault tolerance method for continuing services in mobile computing environment, is researched in this paper. Based on user access patterns, mainly considering the visit time of the mobile host(MH) to the sojourn mobile support stations (MSSs), a checkpointing data storage scheme is proposed. Only if the stay time in the current mobile support station is long enough, the mobile host should do the checkpoint and store the checkpoint data on the sojourn mobile support station. Based on the visited time to the MSS, the scheme manages checkpointing data high efficiently which is more cognitively flexible and adapt to the nomadic and mobile computing environmental conditions.

A Comparison of Small Area Estimation and Kriging in Estimating Rainfall in Sri Lanka

Accurate prediction of rainfall is crucial for a country like Sri Lanka as its economy is mainly based on agriculture. Many statistical techniques have been used in the past to estimate rainfall. The main purpose of this study is to evaluate and compare the accuracy of two techniques in estimating seasonal rainfall in Sri Lanka. The two techniques that were compared are namely; Small Area Estimation and Kriging. The four rainfall seasons considered is First Inter Monsoon, South West Monsoon, Second Inter Monsoon and North East Monsoon. Monthly rainfall data collected at 75 rain gauges during the year 2011 were used in this study. Data is withheld from 14 stations were then used to compare the predictions of seasonal rainfall using both techniques. Root Mean Squared Error, Correlation coefficient and scatter plots of observed and fitted seasonal rainfall values were used to compare the two techniques. The Comparison revealed that Kriging provided better predictions for rainfall in First Inter Monsoon and South West Monsoon. Predictions of both techniques were not much successful in estimating rainfall in Second Inter Monsoon. Small area estimation yielded more accurate predictions for rainfall in North East Monsoon.

Use of artificial neural networks and support vector machines to predict lacking traffic survey data

The aim of this paper was to predict lacking data from a traffic survey along a principal highway in Bangladesh using artificial neural network (ANN) combined with the support vector machine (SVM). Traffic data were obtained at an hourly rate using a methodical inquiry over a four-year period at the Jamuna toll collection point, which is located along the North Bengal corridor of Bangladesh. Two evolutionary computational statistical procedures were used along with its corresponding numerical model. The neural network and SVM were fed with data from 13 recurring weeks over a fouryear period. The missing data were predicted with significant accuracy using both methods. Accuracy of the methods was compared, which showed that the SVM method is much more accurate than the ANN technique. Combination of both the ANN and SVM models can be used to obtain trends in traffic data more accurately.

An Adaptive Controller of Traffic Lights using Genetic Algorithms

In this modern era of technology, time and energy are the most valuable assets of current civil societies. Due to the growing population and number of vehicles on the roads, traffic congestion is becoming an escalating complex problem, and this problem has now grown into a major issue for the urban planning authorities. Therefore, many researchers continue investigating this problem and exploiting modern science to develop reliable mathematical models, simulation scenarios and solution approaches for the traffic light control system that could optimally coordinate the traffic signals to time and reduce traffic congestion. In this paper, we propose a dynamic application that simulates many traffic light models using a Genetic algorithm. We have used the Genetic algorithm for traffic signals time management and to compute optimal solutions for cycle times, offset times and green times according to the sequence orders of a set of traffic lights. We employed the fitness function in order to minimize the time waste on the model. The experimental prototype delivers sound results and adequate optimal solutions.