MERCon - 2018

Permanent URI for this collectionhttp://192.248.9.226/handle/123/14699

Browse

Now showing 1 - 20 of 107

item:
2018 Moratuwa Engineering Research Conference (MERCon) [Pre Text]
(IEEE, 2018)
item: Conference-Full-text
Q-learning approach for load-balancing in software defined networks
(IEEE, 2018-05) Tennakoon, D; Karunarathna, S; Udugama, B; Chathuranga, D
In this paper, we propose a Q-Learning approach for load balancing in Software Defined Networks to reduce the number of Unsatisfied Users in a 5G network. This solution integrates Q-Learning techniques with a fairness function to improve the user experience at peak traffic conditions. With typical high rates offered by 5G and future networks single user behavior shall have a significant impact on the Quality of Service (QoS) on the rest of the users. Therefore, we are in need of responsive networks based on their utilization and on the number of users occupied. In this paper we classify users into different groups and normalize the resources to provide the best QoS. The simulation results verify the improvement in terms of the number of Unsatisfied Users and of the connections dropped. Additionally, it enhances per-flow resource allocation while avoiding over-utilization of certain network resources. In a nutshell, this proposal will serve any future network with high traffic conditions to deliver the best QoS to their end users.
item: Conference-Full-text
Non-invasive blood glucose monitoring using a hybrid technique
(IEEE, 2018-05) Nanayakkara, ND; Munasingha, SC; Ruwanpathirana, GP; Chathuranga, D
Diabetes needs regular blood glucose monitoring to control it. Invasive blood glucose measuring is the current gold standard. It causes discomfort for the patient and sometimes even infections. Researchers around the world have reported different techniques to measure blood glucose levels non-invasively, but a universally acceptable method with required accuracy is not yet available. We proposed a novel approach to measure blood glucose level non-invasively using a hybrid technique combining Near InfraRed (NIR) absorption and bio-impedance measurements. We tested the methods individually first. Then Artificial Neural Network (ANN) and least squares regression were used to integrate the two methods. The combined methods showed better accuracy compared to the individual measurements. The hybrid technique developed using the linear regression models showed a superior outcome with 90% and 10% of the data points in the regions A and B of the Clarke error grid, which are considered acceptable.
item: Conference-Full-text
Distinguishing real web crawlers from fakes: googlebot example
(IEEE, 2018-05) Algiryage, N; Dias, G; Jayasena, S; Chathuranga, D
Web crawlers are programs or automated scripts that scan web pages methodically to create indexes. Search engines such as Google, Bing use crawlers in order to provide web surfers with relevant information. Today there are also many crawlers that impersonate well-known web crawlers. For example, it has been observed that Google’s Googlebot crawler is impersonated to a high degree. This raises ethical and security concerns as they can potentially be used for malicious purposes. In this paper, we present an effective methodology to detect fake Googlebot crawlers by analyzing web access logs. We propose using Markov chain models to learn profiles of real and fake Googlebots based on their patterns of web resource access sequences. We have calculated log-odds ratios for a given set of crawler sessions and our results show that the higher the log-odds score, the higher the probability that a given sequence comes from the real Googlebot. Experimental results show, at a threshold log-odds score we can distinguish the real Googlebot from the fake.
item: Conference-Full-text
Total quality management implementation in Sri Lankan construction industry
(IEEE, 2018-05) Wickramarachchi, HR; Sandanayake, YG; Ekanayake, BJ; Chathuranga, D
Achieving quality in construction projects is vital for construction organisations. Small and medium sized enterprises (SMEs) require vigorous quality achieving techniques to compete with large scale competitors. Therefore, this study investigated the total quality management (TQM) implementation in construction SMEs in Sri Lanka. The study adopted a mixed approach. Data collection was done through preliminary interviews and a questionnaire survey among professionals of construction SMEs. Quantitative data were analysed using respondent average score method and qualitative data were analysed using content analysis. The study revealed that although TQM practices are employed by the SME contractors, their understanding about TQM is low. The highest implemented TQM practice is the “Continual review and comparison of work progress and cash flow with data related to actual progress”, while “Conducting value engineering workshops with client at project commencement” is the least implemented. TQM practices of “Employees feel the commitment of top management towards project and process quality improvement” and “Ensuring prompt communication of design changes to the project team and other relevant parties during the construction process” are the most contributor to project success. “Employees are given with responsibility and authority to perform without monitoring and inspections”, is the least contributor to project success.
item: Conference-Full-text
A modified test statistic for maximum-minimum eigenvalue detection based on asymptotic distribution thresholds
(IEEE, 2018-05) Kartchner, DR; Jayaweera, SK; Chathuranga, D
The ratio of the maximum and minimum eigenvalues of the sample covariance matrix has been suggested as a test statistic for signal detection in low-SNR regimes. The threshold required to implement a Neyman-Pearson test on this statistic is usually computed by estimating the distribution of this eigenvalue ratio under the null hypothesis using results from random matrix theory (RMT). However, in order to apply asymptotic laws from RMT, the data matrix used to construct the test statistic must have statistically independent columns, which was not satisfied by the test statistics used in previously proposed detectors. This paper forms a data matrix with independent columns to compute the test statistic for maximum-minimum eigenvalue (MME) detection and compares its performance to that of the test statistic as currently defined in literature. The comparison is made with both the semi-asymptotic threshold, which uses the limiting distribution of the maximum eigenvalue and the asymptotic constant to which the minimum eigenvalue converges; as well as the limiting distribution-based threshold, which uses the limiting distribution of the ratio of the maximum and minimum eigenvalues. Simulations compare the expected false alarm rate versus actual false alarm rate, as well as the receiver operating characteristic (ROC) for the following three cases: the two test statistics with the semi-asymptotic threshold, the two test statistics with the limiting distribution threshold, and the two thresholds in conjunction with the newly proposed test statistic. Results demonstrate that the newly proposed test statistic with the limiting distribution threshold is the only case where the actual false alarm rate remains consistently below the false alarm constraint set in the Neyman-Pearson test, while the previous test statistics are almost completely unresponsive to changes to the false alarm constraint.
item: Conference-Full-text
Simulation of online human arm inertia estimation for robot-aided rehabilitation
(IEEE, 2018-05) Harischandra, PAD; Harsha, AM; Abeykoon, S; Chathuranga, D
Most of the studies on rehabilitation robots consider the human arm inertia and the gravity torque as system disturbances. Individual anthropometry varies from patient to patient, and therefore human limbs are not modelled. Some studies used the Disturbance Observer (DOB) as a method of disturbance rejection. However, if the inertia and gravity torque parameters of the human arm could be estimated, they could be effectively used in the controller loop to achieve precise motion control. This paper proposes a novel Reaction Torque Observer (RTOB) based estimation technique which updates parameters using learning and recursive algorithms in real-time. The proposed method is applicable to many robot systems where the load inertia or the load is not known. A simulation was carried out with realistic parameters to compare the performance of two competing methods proposed namely, Adaptive Linear Neuron (ADALINE) and Recursive Least Squares (RLS). Results show that the RLS method outperforms the ADALINE method based on the performance criteria of accuracy, precision and convergence speed for estimating the inertia.
item: Conference-Full-text
Topological structure of the road traffic network in the western regional Megapolis, Sri Lanka
(IEEE, 2018-05) Perera, S; Perera, HN; Perera, K; Chathuranga, D
Recent advances in network science have encouraged researchers to adopt a topological perspective in network assessment. This represents a paradigm shift from traditional network analyses for which the primary focus was flows. In this paper, we present a comprehensive topological investigation of the Road Traffic Network (RTN) of the Western Regional Megapolis in Sri Lanka. We identify the critical nodes (intersections) and links (road sections) within the RTN based on various network theoretic metrics. In addition, we assess the topological robustness of the RTN based on sequential failure (both random and targeted based on connectivity) of the intersections in the system. Our findings could inform policymakers in prioritizing the scarce funds available for future infrastructure investments.
item: Conference-Full-text
Embodied carbon and cost analysis to identify the most appropriate wall materials for buildings: whole life cycle approach
(IEEE, 2018-05) Anuradha, IGN; Perera, BAKS; Mallawarachchi, H; Chathuranga, D
Construction industry is one of the main exploiters of natural resources. During building construction, a large amount of Embodied Carbon is released to the atmosphere. Thus, this research was aimed to determine the significance of whole life Embodied Carbon (EC) in the selection of wall materials for use in building construction. To achieve this aim, mixed research approach was used. A comprehensive literature review and ten expert interviews were conducted first as a part to find out the significance of EC to the construction industry. Subsequently, a questionnaire survey and a base model building consisting of work studies, a desk study and a documentary review were used to collect the required data. The data collected were analyzed to find out the most suitable material out of the three types of wall materials selected. The results of the analysis indicate that eco interlocking brick is the most suitable wall material (compared to cement block and conventional brick) in terms of EC and initial and maintenance cost. The study also provides a firm base for future researchers engaged in the studies to find out the EC content of building materials.
item: Conference-Full-text
Dangers of component trapping in distillation: an industrial methanol distillation case study
(IEEE, 2018-05) Udugama, IA; Mansouri, SS; Kirkpatrick, R; Young, B; Taube, MA; Chathuranga, D
The formation of organic acids due to secondary reactions is an issue in industrial methanol synthesis. As such, to avoid the formation of acidic regions in the units downstream of the methanol synthesis loop, caustic dosing is a common practice in the industry. Despite these precautions, some organic acids can be left in crude methanol discharge from the methanol synthesis loop. The objective of this study was to identify if the mode of operations in the methanol distillation units that purifies the crude methanol into high purity product can potentially lead to an accumulation of trace organic acids within the main refining column, which can lead to the formation of an acidic region within the column. To carry out this work, the main refining column of an industrial methanol producer was first simulated on the industrial process simulation platform VMGSim, and then validated against available data. This simulation was then used to study the accumulation of organic acids ranging from formic acid to valeric acid, where they were added to the feed stream at a concentration of 1 ppm. The study found that propionic and butyric acid, in particular, can accumulate significantly in the middle of the column reaching a concentration of 40 to 80 ppm creating an acidic environment ( PH 3.63), which can cause corrosive damage.
item: Conference-Full-text
Impact analysis of us dollar index volatility on imports and import categories of Sri Lanka
(IEEE, 2018-05) Sahabandu, RV; Asanka, PPGD; Chathuranga, D
The economic liberation in 1977 resulted in drastic changes in many aspects of Sri Lanka. Considering about 1978- 2015, the country yearly import demand represents over 30% share of the gross domestic product (GDP) except 1984, 2009, 2010, 2013–2015. Investigations and the studies on a countries’ imports are surprisingly overlooked as there are several studies being carried out focusing only the aggregated export volume concerning the exchange rate volatility. The monthly data of Sri Lanka imports, import categories and monthly US Dollar (USD) volatility from January 2007-December 2016 were used for the analysis. This study tries to learn the impact of US Dollar Index (USDX) volatility on import demand of Sri Lanka. The Autoregressive Distributed Lag (ARDL) Approach is employed to learn long-term and short-term cointegration among the underlying variables. There exists a 95% statistically significant short-run relationship and it is identified that the import categories, Consumer Goods (CG), Intermediate Goods (IG), Investment Goods (INV), Unclassified Items (UI), None-Oil Imports (NO) have a speed of adjustment to the equilibrium (SAE) in the long-run of 17%, 36%, 23%, 23%, 25% respectively. The total imports reveal that the disequilibrium conditions will be resolved by 27% within a period of one month that is shocked due to the USDX volatility. Knowledge of the relationship between USDX fluctuation, exchange rate volatility and import volume will support to pursuit for a beneficial trade and prevent or be prepared for a much more stable situation within Sri Lanka.
item: Conference-Full-text
An in-depth study of ssvep signals against stimulus frequency and distance to the stimulus
(IEEE, 2018-05) Pathiranage, S; Paranawithana, I; Perera, M; De Silva, AC
In recent times, Steady State Visually Evoked Potentials (SSVEP) based BCI have gained popularity over different types of brain signals as they prove to demonstrate interesting results. It is important to understand the behavior of brain signals with the change of frequency and distance to the stimuli used to evoke them. In this paper, we have looked at the behavior of the SSVEP signals over 4 varying stimuli frequencies and 4 varying distances between the subject and the stimulus. It was found that the strongest SSVEP were elicited in the lower frequency range around 8Hz and a distance up to 100 cm can have a significant effect on the elicited SSVEP signals.
item: Conference-Full-text
Beyond site material handling and transportation in large-scale construction projects
(IEEE, 2018-05) Abeysinghe, AMGP; Waidyasekara, KGAS; Melagoda, DG; Chathuranga, D
There is a significant contribution of materials to be handled, transferred, and used; further, large logistics operations are related to transporting, storing, moving, and reloading in large scale construction projects. Thus, the research aims to investigate issues that occur beyond the site, material handling, and transportation in large-scale construction projects. Accordingly, extensive literature was synthesized to identify the usage of materials in large-scale construction projects, their significance, and procedures of material handling and transportation. This study adopted a qualitative research approach by selecting a multiple case study strategy. Five largescale construction projects in Colombo Metropolitan area were selected and semi-structured interviews were conducted with three different levels of site personnel from each case study. Finally, the gathered information was analyzed using computerbased content analysis and further supported by data collected through observations. The outcomes disclosed various issues in material handling and transportation with respective causes and possible solutions to minimize identified issues in large scale construction projects. Accidents, adverse weather conditions, lack of material handling equipment, lack of labors, improper packing systems, misplace and steal of materials, lack of pre-arrangements, bulk quantities, limited site area, delay in taking approvals, improper supervision and unawareness of the handling process were identified as common material handling and transportation issues. It was identified these can be mitigated by proper supervision, providing a manual to handle each material, labeling material, maintaining good communication with laborers and arranging a gate pass system. Further, recommendations for improving material handling and transportation were provided for industry practitioners.
item: Conference-Full-text
Development of a quad-rotor fixed-wing hybrid unmanned aerial vehicle
(IEEE, 2018-05) Gunarathna, JK; Munasinghe, R; Chathuranga, D
This paper describes the development of a quadrotor fixed-wing hybrid unmanned aerial vehicle (UAV). By combining the flying qualities of quadrotor UAV and that of fixed-wing UAVs it is possible to achieve superior flying qualities. This way, the vertical take-off and landing capability of quadrotors and the long endurance of fixedwing UAVs have been put together, while eliminating the disadvantages of both of these UAV types. The proposed hybrid drone has shown exciting performance in takingoff, cruising and landing fully autonomously. Control transition from quadrotor to fixed-wing and backwards have been analyzed. Sky Scout fixed-wing drone has been used to build the hybrid UAV, and the performance has been successfully demonstrated through flight tests. This paper precent the complete design of a quadrotor fixed-wing hybrid aerial vehicle with modelling, simulation and experimental verification.
item: Conference-Full-text
Feature oriented software development framework for stock exchange systems
(IEEE, 2018-05) Konara, L; Perera, I; Gunewardhane, S; Chathuranga, D
Many software companies use the popular method of layered architecture to develop their software products resulting in code to be more generic. This generalization introduces a lot of unnecessary elements which make the system architecture to be heavy and less elegant. To overcome this challenge, a concept of feature oriented software development (FOSD) is introduced. That way, the software system will be delivered as a set of features and the feature set could be decoupled at any time, according to the latest set of requirements without harming to any existing functionality. In this research the final outcome has been a feature oriented domain specific language (DSL). The objective of this developed feature oriented DSL is to make it effective even for business analysts to introduce new features without getting help from the development team of the project. A code generator and a core controller have been developed as part of this research exhibiting the above mentioned FOSD & DSL principles. The evaluation results indicate the generated feature oriented code identically behaves as its manually written counterpart code.
item: Conference-Full-text
A study on natural rain surface erosion of different walling materials in tropics
(IEEE, 2018-05) Udawattha, CD; Galkanda, GAHH; Halwatura, RU; Chathuranga, D
This study presents most common phenomena observed on wall facades due to excessive rain in tropical climatic condition in Sri Lanka. The objective of this study is to understand the effect surface decay of walling materials. The most common walling materials such as brick, cement block, earth cement blocks and novel walling materials mud concrete block, geopolymer blocks were subjected to this study. Faade covering materials such as rough cement plaster and cement slurry plaster were studied. Intrinsic material properties were studied prior to natural rain erosion quantification. Basic engineering properties of water absorption, capillary action, surface roughness, compressive strength was studied. Material properties such as water absorption, surface roughness were studied. Ten years of rain was simulated to measure natural surface decay. The results show that, stronger materials and less surface rough materials are resistant to surface decay. Plastering is recommended to protect walling materials reduce the surface decay. Comparatively stronger walling materials are less prone to natural rain decay. Cement plasters and cement slurry plaster covering is the most suitable materials for heavy rain. Cement slurry plaster is the most suitable walling covering for rain drop bouncing areas in outdoor walls in Sri Lanka.
item: Conference-Full-text
Analysis of the dimensional change of woven fabrics from loom state to finished state
(IEEE, 2018-05) Kandemulla, KASM; Maduwantha, ARL; Fernando, EASK; Niles, SN; Jayawardana, TSS; Chathuranga, D
Dimensional change of a woven fabric is a challenge in woven fabric production. The phenomenon originates from the dimensional instability of the fabric. Shrinkage is a combined result of numerous factors such as relaxation, dyeing, finishing and the effect of machinery. The significance of this problem has been investigated by several researchers, who focused mainly on the geometry of the fabric during the weaving phase (loom stage) only. To investigate the dimensional changes that occur within a particular woven structure, a number of fabric samples were sent through various finishing processes such as, washing, dyeing & finishing, under the laboratory conditions. The changes were measured either in terms of dimensional change or EPI and PPI values. Using the experimental data and theoretical analysis, a mathematical model has been developed and validated. However initially the focus is laid on plain woven fabrics and it is expected to be further extended to the other woven structures as well.
item: Conference-Full-text
Evaluation of different classifiers for sinhala pos tagging
(IEEE, 2018-05) Fernando, S; Ranathunga, S; Chathuranga, D
This paper presents a comparative evaluation of three state-of-the-art classifiers for Sinhala Parts-of-Speech (POS) tagging. Support Vector Machines (SVM), Hidden Markov Models (HMM) and Conditional Random Fields (CRF) based POS tagger models are generated and tested using different combinations of a corpus of news articles and a corpus of official government documents. CRF is used for the first time in Sinhala POS tagging, thus the best feature set is experimentally derived. To further improve the accuracy of POS tagging, a majority voting based ensemble tagger is created using three individual taggers. This ensemble tagger achieved the highest accuracy in POS tagging than any individual tagger. The two domains (news, and official government documents) used in this study have noticeable differences in writing style and vocabulary. Generating domain specific POS taggers is time consuming and costly due to the overhead involved in creating and manually tagging domain specific corpora, for low resourced languages in particular. Therefore, this study also evaluates the possibility and successfulness of using corpora of different domains in training and testing phases of aforementioned machine learning techniques.
item: Conference-Full-text
Development and characterization of a soft tactile sensor array used for parallel grippers
(IEEE, 2018-05) Weerasinghe, L; Chathuranga, DS; Chathuranga, D
Tactile Sensors play a crucial role in developing robots with human like grasping and manipulation capabilities. Localization, dynamic sensing and good force recognition characteristics are major goals when designing a tactile sensor. However, fulfilling these requirements come at the cost of increased complexity in design, high cost and difficulties in practical implementation due to size. In this research a sensor has been developed that is based on the concept of Hall effect. An array of magnets and hall sensors create a unique combination of outputs for each different deformation of the dual layered silicon membrane which houses the magnets. While allowing the interaction with non-planar surfaces due to the compliant nature of the silicon material, the sensor also facilitates accurate force recognition and localization with super-resolution using sensor readings, geometry and elastic properties of the silicon layer. This paper contains the design, fabrication and calibration of the tactile sensor array.
item: Conference-Full-text
Development of driving cycles for Galle
(IEEE, 2018-05) Rupasinghe, H; Rengarasu, TM; Chathuranga, D
Driving cycle construction has become more popular around the world to estimate pollution emission and traffic analysis. This paper presents the traffic based analysis related to performance of B128 and B130 route in Galle, Sri Lanka using basic traffic engineering parameters such as travel time, average and maximum speed, average acceleration and deceleration. So the main objective of this paper is to identify traffic related parameters of above routes at Galle. The driving data was collected using on board measurement method by installing a prototype GPS tracking device in light-duty cars (AXIO, AQUA, FIT, Vitz). The cycle construction was done according to the segment based cycle construction method and roads were segmented considering the major intersections. Driving cycles were constructed for morning peak, noon peak, evening peak and off peak hours in B130 route and off peak hours in B128 route and the travel time, average speed, maximum speed, average acceleration and average deceleration were obtained. According to results, the maximum average speed of 27.53 km/h was obtained at off peak hours and travel time of 892s is taken to travel from Hapugala junction to Galle in noon peak hours. However the maximum speed was obtained at noon peak hours.

Browse

Recent Submissions