Dileeka Dias
University of Moratuwa
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Dileeka Dias.
international conference on information and automation | 2008
B.D.S. Lakmali; Dileeka Dias
Estimation of the location of a Mobile Station accurately is a key requirement to effectively provide a wide range of Location Based Services over mobile networks. Hence developing cellular positioning techniques has been a key research problem, with numerous localization solutions been proposed. These include technologies such as Cell ID, angle and time of arrival methods, statistical methods and fingerprinting methods. This paper presents fingerprinting based positioning techniques suitable for outdoor and indoor positioning. Multiple positioning techniques were proposed, implemented and evaluated with outdoor and indoor trials. The ultimate solution proposed in this paper is not a single positioning technique; rather it presents several positioning techniques that achieve optimum performance in each test environment. The results reports 67% positioning error as 112 m, 299 m and 221 m for urban, suburban and rural areas respectively. Experimental results show that the proposed positioning methods achieve accuracy far better than Cell-ID and trilateration approaches for the tested network environments especially for rural areas. The 67% positioning error for rural area is 1045 m and 1386 m with basic Cell-ID and trilateration techniques while proposed fingerprinting based technique reports 67% positioning error as 221m. With indoor positioning this paper reports 50% positioning error as 8.7 m and also it was possible to accurately differentiate between floors in the selected multi storey building.
wireless and optical communications networks | 2010
N. S. Kodippili; Dileeka Dias
In this paper, a new algorithm for indoor localization using WLAN signal strengths is proposed. The objective of the work was to design a simple and efficient indoor positioning algorithm with good accuracy. The proposed algorithm uses the fingerprinting technique as a pre-processing step. The second step is the application of trilateration within the nearest neighbours found during the first step. The paper describes the algorithm, the environment used to test the algorithm and results obtained. The results show that the proposed algorithm has 44% better accuracy compared to basic fingerprinting technique and the accuracy is 73% better compared to basic trilateration technique. The algorithm has been demonstrated to be independent of the propagation environment and works with no knowledge of the positions and the transmit powers of the WLAN access points in the vicinity. Thus, it is a simple and effective indoor positioning scheme.
international conference on spatial data mining and geographical knowledge services | 2011
Kushani Perera; Dileeka Dias
This paper suggests a decision support system for vehicle drivers, accessible via mobile phone. Concept behind the system is to help drivers to schedule their activities, best utilizing their time along the way, minimizing the impacts of traffic. This is in contrast to existing approaches focused on controlling traffic in highways. Vehicle is continuously tracked along the journey and information presented to user is adapted according his location and time dimensions. System is based on a decision tree based classification model to predict the future traffic and use those results for decision making. System mines spatio-temporal data to build the decision tree, therefore developed in a distributed architecture to avoid load for a single server. System is exposed to community via existing social networks, bringing social networks into vehicular context. Decision trees to predict traffic are periodically rebuilt using most recent data, therefore this is an intelligent system which learns through empirical data, and best suited for dynamic vehicular environment.
global humanitarian technology conference | 2011
Lanka Wijesinghe; Prasanga Siriwardena; Shamali Dahanayake; Dharshana Kasthuriratne; Ravi Corea; Dileeka Dias
We present the design and implementation of intrusion detection and alerting mechanism (eleAlert) for fences separating wildlife habitats and human settlements. Our objective is to improve the effectiveness of electric fences as a solution to the prevailing human-elephant conflict (HEC) in many parts of the world. eleAlert uses a network of sensors to detect and locate damages instantly and alert communities under threat via the mobile communications network.
international conference on industrial and information systems | 2009
H. A. N. C. Bandara; Dileeka Dias
This paper proposes a novel method for a dynamic ridesharing mechanism using the combination of several modern information and communication technologies.
international conference on information and automation | 2008
Pushpika Wijesinghe; Dileeka Dias
The database correlation method (DCM) is a network based positioning technology which has shown superior in terms of accuracy. DCM is based on a pre-measured database of a location dependent variable such as received signal strength (RSS). Even though the technique has good potential, the practical difficulty in forming the database (fingerprints) using field measurements has become the major challenge in implementing this in large, dynamic networks. A remedy for this is to make use of propagation model predictions instead of field measurements to create the fingerprints. However, due to the considerable deviation between the predictions and the actual measurements, the positioning accuracy diminishes with this approach. In order to overcome this issue, tuning of the predictions using a small number of field measurements can be applied. The work presented in this paper proposes a technique for the correction of such deviations which would improve the performance of DCM. The proposed tuning process, cell-wise calibration, is based on artificial neural networks (ANN). Two different training algorithms, particle swarm optimization algorithm (PSO) and BFGS algorithm are applied for ANN training. The results of the trials carried out in urban, suburban and rural environments are presented. With the PSO algorithm, the level of accuracy is comparable to that obtained with a measured fingerprint database in urban and suburban environments, and is better in rural environment.
international conference on information and automation | 2010
Deneth Karunarathne; Thanushka Gunasekara; Dileeka Dias; Dharshana Kasthurirathne
We present an effective solution for both digital map generation and real time user tracking for teams operating in hazardous environments or unfamiliar territory. The system is based on the client server model, where the client is a GPS enabled mobile phone which is capable of capturing coordinates and serving dynamic maps to the user. The maps plot the teams navigational routes and enables real time tracking of each other. The server builds a GIS at the back end based on the data sent by the users, and also develops maps on demand. The system is implemented with simple hardware and uses the available GPRS network for data gathering. This enables the system to be used both in hazardous environments and by the general public.
international conference on industrial and information systems | 2009
Michael Medagama; Dileeka Dias; Shantha Fernando
Bandwidth scarcity is a common problem faced in multimedia transmission over broadband wireless networks. From a consumers perspective, the effectiveness of multimedia delivery suffers significantly in such networks. Further, in a mobile environment, when the number of users increases, they experience interruptions in the received multimedia streams. Adaptive streaming is one solution that can be adopted in any network environment regardless of the nature of the network. Adaptive streaming can be achieved using transcoding techniques. Transcoding is a mechanism that converts a video to a form that has less information so that the resulting data volume is appropriate for streaming over a low bandwidth network. However, the quality of the video stream drops due to transcoding. The advantage is that when there is a large number of users sharing the network, users could watch a video at a lower quality with no interruption if the transcoding is applied before the video is streamed to the end user. i.e., a compromise can be made between the video quality and network delay. The main focus of this research is to analyze the variation of transcoding parameters with respect to bandwidth, in order to achieve an adaptive streaming solution of optimum achievable quality in a low bandwidth broadband network.
international conference on industrial and information systems | 2009
Dasuni Kannangara; Nimalika Fernando; Dileeka Dias
A web enabled visualization methodology is proposed in this work for viewing time varying spatial data. The technique is simple, and hence could be used by non experts of IT or GIS. The user can upload own datasets prepared in a given format with specific fields. The variation of time for data visualization can be adjusted. A web server and centralized GIS engine would process the data and the visualization would be sent to the web client as an animation. Spatial and temporal interpolation is used at the server side. In order to test the methodology, a GIS based Time-varying Spatial Information Visualizer is developed. The methodology is tested with local weather data. It can be used for visualization of demographic data as well. The methodology can be further developed to add forecasting facilities based on user provided data.
joint ifip wireless and mobile networking conference | 2013
Yogatheesan Varatharajah; Nuwan Karunathilaka; Mohamed Rismi; Sujan Kotinkaduwa; Dileeka Dias
The paper presents a wireless body area sensor network (BASN) for evaluating the effectiveness of fitness exercising. The goal of our work is to provide a mobile and an accurate way of quantifying the energy expenditure during a fitness exercise session. While energy expenditure estimates are available in modern exercise machines, no accurate method exists for outdoor exercising. The prototype system implemented includes an IEEE 802.15.4(Zigbee) based wireless body area sensor network which measures physiological and kinetic features related to the activity being monitored, and a mobile application for visualization of the estimated energy expenditure. The paper outlines the sensor hardware used, the body area network which collects the measurements from the sensor nodes, the algorithm for energy computation and experimental results. The system, once trained can be used with any form of outdoor exercise. The key contribution and novelty of the paper is the development of a mechanism to establish the hitherto unknown relationship between physiological and kinetic features and body metabolism through the BASN. Further, this relationship may be personalized to individual level with proper training.