Browsing by Author "Seneviratne, J. A."

Now showing 1 - 10 of 10

A cost-effective and adaptable queue management system to increase efficiency in patient queue management
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Adhikari, A. M. N. D. S.; Gunarathna, T. G. L.; Bandara, K. D. Y.; Gunawardana, K. D. B. H.; Seneviratne, J. A.; Perera, M. H. M. T. S.
Healthcare systems worldwide, particularly in resource-limited settings like Sri Lanka, face significant challenges related to high patient volumes and constrained resources. These challenges often lead to extended wait times and reduced patient satisfaction. This study presents an innovative, adaptable queue management system designed to replace inefficient manual methods, enhance operational efficiency, and optimise patient flow. Scalable to meet the needs of both small clinics and large hospitals, the system functions across various connectivity scenarios, ensuring flexibility in diverse environments. The system comprises patient, doctor, and administrative interfaces. Upon patient registration, a QR code will be generated, and the patient can use the QR code to check-in. A printed queue token will be issued when a patient checks-in. Doctors can manage their queues and access real-time patient information. Administrators oversee overall system operations, including advertisement management and key performance indicator (KPI) tracking, to monitor and enhance healthcare delivery in addition to having the ability to add, remove, or edit users. Built on a robust technology stack that includes HTML, CSS, JavaScript, PHP, SQLite3 for database management, and AES-256-CBC encryption for secure data handling, the system is designed for reliability and scalability. Embedded ESP32 devices with OLED displays and LEDs provide offline functionality, while multicast DNS (mDNS) ensures seamless device connectivity to local networks without requiring Internet access which is critical for rural healthcare facilities. The system features a custom-built algorithm, leveraging Random Forest Regression, to analyse historical and real-time queue data. This allows for precise queue time estimates and significantly improves staff and patient planning. The system outperforms the traditional manual systems, which lack both real-time prediction capabilities and efficiency. The system performance was meticulously improved using various optimisation techniques such as batch processing, database indexing, and algorithm optimisation, which led to an execution time of 22 seconds to be brought down to 1.5 seconds on a 1.4 million row data set, where the execution involved processing, sorting, encrypting, decrypting, and storing data. A one-tailed t-test was performed to compare the execution times of test runs with optimisation and without optimisation. There was a significant difference in execution times between test runs without optimization (M = 21.84, SD = 1.16) and execution times between test runs with optimization (M = 1.52, SD = 0.28); t(43) = 107.76, p < 0.001. The system was validated for 10 years of sample data and the results demonstrate that the system is robust and responsive under real-world conditions. Continuous validation is ongoing in diverse healthcare environments to further assess its impact on optimizing queue management, resource allocation, and patient satisfaction. This scalable and adaptable system represents a substantial advancement in healthcare management, offering a transformative solution to meet the evolving needs of healthcare facilities despite scarce infrastructure.
Design and characterization of a tunable metamaterial absorber for efficient RF energy harvesting in the Wi-Fi frequency band
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Sahasrika, S. L. S. D.; Ranaweera, A. L. A. K.; Seneviratne, J. A.; Jayathilaka, K. M. D. C.
The pursuit of miniaturization and connectivity in semiconductor technology has led to the development of compact, interconnected devices. However, the reliance on batteries for power presents limitations. Ambient energy harvesting, particularly from radio frequency (RF) waves, offers a promising solution. This study explores the feasibility of wireless energy harvesting, especially in the context of the saturated frequency spectrum due to wireless communications. Building upon recent advancements in metamaterial technology, this study focuses on the design, fabrication, and characterization of a tunable metamaterial absorber unit cell for efficient RF energy harvesting. The aim is to exploit metamaterials, specifically tunable metamaterial absorbers (MMA), to harvest RF energy, particularly in the widely utilized Wi-Fi frequency band. To design a novel unit cell structure, simulations were conducted using the commercially available EM simulation tool CST Microwave Studio software. A novel MMA unit cell consisting of E-shaped split ring resonators, copper reflector, and FR-4 substrate layer in the middle was designed and simulated. Results demonstrated its RF energy harnessing ability, achieving peak absorptivity of 98% of incident RF energy at 2.4 GHz. The designed unit cell was fabricated using the standard PCB fabrication method. Experimental results were obtained using a network analyzer through non-contact measurement method. Results closely mirrored the simulation results, confirming a high absorptivity of 99% at 1.44 GHz. To achieve frequency tunability, an external capacitor switching circuit was integrated into the MMA unit cell. Simulation and experimental results were obtained confirming its frequency tunability. The proposed tunable metamaterial absorber unit cell offers advantages over conventional RF energy harvesting systems, including ease of implementation, a wider range of RF energy absorption, and cost-effectiveness paving the way for integration into various applications. In conclusion, this study contributes to the development of energy harvesting technologies by leveraging tunable MMA to harness ambient RF energy.
Design of novel perfect metamaterial absorber for Radio Frequency energy harnessing
(Faculty of Science, University of Kelaniya Sri Lanka, 2022) Karunathilaka, M. G. M. T.; Ranaweera, A. L. A. K.; Jayathilaka, K. M. D. C.; Seneviratne, J. A.
Due to the rapidly growing wireless communications and sensing applications, the frequency spectrum has already been saturated. Consequently, the abundance of Radio Frequency (RF) signals in the ambient environment made the concept of wireless energy harnessing to be emerged as an attractive solution to energize low-power wireless devices. In this study, a novel tuneable perfect metamaterial absorber (PMA) unit cell was designed by combining two C-shaped split-ring resonators (SRR) embedded with simple electronics circuitry. The feasibility of harnessing energy from 1.8 GHz signals was investigated through electromagnetic (EM) simulations. The design and numerical analysis of the proposed PMA structure is carried out with the aid of the commercially available EM simulation software, High Frequency Structure Simulator (HFSS). The proposed structure’s capability to absorb EM energy as a perfect metamaterial absorber is studied. According to the simulation results, it shows a high absorption coefficient of around 99%. This verifies that the proposed tuneable PMA structure encompasses a high absorption of RF energy. It can be used for the harnessing of RF energy to power up low-power devices and wireless sensor networks.
Designing a high-performance parametric speaker system: simulation and Optimization
(Faculty of Science, University of Kelaniya Sri Lanka, 2023) Gurusinghe, T. N.; Seneviratne, J. A.; Ranaweera, A. L. A. K.; Kalingamudali, S. R. D.
The parametric speaker is designed to direct omnidirectional sound waves towards a specific target. Over the past two decades, numerous research studies have been conducted to optimize parametric speaker systems, with a focus on enhancing audio quality and extending the range of sound propagation. The objectives of this research include the enhancement of the audio quality, the reduction of total harmonic distortion through modulation techniques, and the amplification of the modulated output to increase the effective hearing distance. These goals were pursued alongside the development of a properly designed ultrasonic transducer array circuit, a critical component of a parametric speaker system. Prior to the conception of the novel parametric speaker system, a comprehensive simulation study was conducted using the commercially available COMSOL Multiphysics software. For this study, the Pressure Acoustics, Frequency Domain (acpr) interface, the Solid Mechanics (solid) interface, and the Electrostatics interface were utilized. The principal aim of this simulation study was to analyse the minimal electric potential required for an ultrasonic transducer element to generate a directional sound wave capable of propagating over a one-metre distance. To achieve this, a PZT-5H piezoelectric element with a stacked aluminium metal diaphragm was constructed. The electric potential across the piezoelectric plate was step by step varied from 5 V to 100 V. Polar plots illustrating the sound pressure level of ultrasound propagation in the air domain at a distance of one-thousand-fourhundred millimetres from the source were generated for each simulation. The simulation model of the piezoelectric element was meticulously constructed after a thorough examination of a crosssectional cut of an ultrasonic transducer and the arrangement of layers within the metal cover. This model adopted a two-dimensional (2D) axially symmetric space dimension. This approach leveraged the rotational symmetry of the elements to simulate in 3D, thereby reducing simulation complexity. The analysis revealed that when the electric potential was below 10 V, the sound pressure remained below 60 dB. However, upon increasing the electric potential to above 60 V, although the expected directionality was achieved, distortions adversely affected the output signal. Such sound propagation characteristics were deemed unsuitable for a parametric speaker system. Upon analysing the polar graphs generated for a 30 V electric potential, it was evident that directionalized sound pressure levels in the air were achieved with minimal distortions compared to other simulated systems. Consequently, a 30 V electric potential was selected as the amplified signal voltage peak-to-peak for application to the designed ultrasound speaker. This approach was undertaken to ensure optimal performance and minimize distortion in the parametric speaker system.
Development of a cost-effective real-time commuter counting system for public transportation in Sri Lanka
(Faculty of Science, University of Kelaniya Sri Lanka, 2023) Premasiri, R. H. M. D.; Koralage, K. G. S. D.; Hasaranga, J. P. K.; Alawaththa, A. K. N. A.; Seneviratne, J. A.; Ranaweera, A. L. A. K.
The use of smart systems in public transportation is relatively new in the Sri Lankan context. This study introduces a cost-effective solution for accurately counting the number of bus passengers at any given time. Current passenger counting systems in Sri Lanka often suffer from inaccuracies and inefficiencies, hindering the effective management of public transportation and addressing problems including long bus queues, ticketing fraud, long waiting times, etc. Precise passenger count is essential for optimising services and resources, ultimately enhancing the efficiency of public transportation in the country. The proposed system is designed to be installed in buses and includes an online platform where users can enter the bus number and check the passenger occupancy in real-time. The proposed system comprises several key components: two input sensors, a microcontroller, a wireless connectivity module, an in-built display, an input panel, and an output display. The two sharp IR sensor modules employed as input sensors are connected to a microcontroller. An inbuilt display connected to the microcontroller provides output information, including passenger count, number of vacant seats, GPS signal strength, and data transmitting capability. Data is wirelessly sent to a cloud database for storage, retrieval, and processing, enabling users to access relevant information via a web application. The algorithm employed in this system ensures precise passenger count by detecting specific sequences of readings from the two IR sensors. To increment the passenger count, the system requires the following sequence: "0-0, 1-0, 1-1, 0-1, 0-0." This sequence corresponds to the detection of a passenger boarding the bus. Each step in the sequence represents the state of the two IR sensors, with "0" indicating no obstacle and "1" indicating an obstacle (i.e., the presence of a passenger). The algorithm recognizes this sequence as an entry event and increments the passenger count accordingly. Conversely, the algorithm relies on the following sequence to accurately decrease the passenger count when a passenger exits the bus: "0-0, 0-1, 1-1, 1-0, 0-0." This sequence represents a passenger leaving the bus. The algorithm reduces the passenger count by monitoring the sensor readings and identifying this sequence. These specific sequences in the algorithm ensure reliable and accurate passenger counting. By requiring a particular order of sensor readings, false positives or negatives caused by noise or temporary obstacles are minimised, leading to a more precise passenger count. The online platform allows users to access passenger occupancy in a particular bus, which aids in real-time service optimization for public transportation management. The system achieves precise real-time passenger occupancy tracking using two Sharp IR sensors and a finely tuned algorithm. Tested results of the pilot system show that data empowered public transportation management in Sri Lanka by optimising routes, efficiently allocating resources, and significantly improving the overall commuter experience. In conclusion, the combination of affordability, reliability, and user-friendliness makes this proposed solution suitable for efficiently managing public transportation systems.
Innovative fence monitoring system to mitigate human-elephant conflict
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Bodaragama, B. T. P.; Athawuda, A. H. C.; Gunawardhana, M. A. W. S. N. T.; Senanayake, S. V.; Leanage, H. B.; Gunawardana, K. D. B. H.; Seneviratne, J. A.
Human-elephant conflict (HEC) poses a significant threat to communities and wildlife, prompting the development of an innovative standalone device to enhance electric fence monitoring and mitigate associated risks. This research introduces a system that determines breakage location by measuring fence capacitance, inductance, and resistance, and uses a mathematical model to map changes of these parameters to change of the fence length. This approach enables remote detection of both open and short circuit breakdowns without relying on expensive, failure-prone active nodes along the entire fence. The device can identify the distance to the breached location along the length of the fence approximately and immediately. It will send this alert via SMS to designated contacts using a GSM module, providing real-time monitoring and rapid response capabilities. This key feature ensures timely alerts and quick responses to potential breaches, enhancing the fence's effectiveness in preventing elephants from entering villages and reducing HEC incidents. The standalone nature of the solution simplifies installation and maintenance, eliminating the need for additional wiring or complex infrastructure, thereby significantly reducing overall costs associated with fence monitoring while increasing reliability and efficiency. Furthermore, the device functions accurately by minimizing the effects of weather changes, ensuring consistent performance in various environmental conditions. This innovative breakage detection system represents a significant advancement in fence monitoring technology for wildlife conservation, addressing many shortcomings of traditional solutions by offering a cost-effective, efficient, and reliable method for mitigating human-elephant conflict. The research underscores the potential of integrating advanced technology with traditional conservation methods to create more sustainable and effective strategies for managing human-wildlife conflicts, ultimately improving the effectiveness of electric fences in deterring elephants and reducing the incidence of fatalities and crop damage. Testing on a 150m fence demonstrated promising results, with the system achieving nearly 80% accuracy in detecting and locating both open circuit and short circuit breakages, as verified through manual simulations and observations recorded in the device's test results.
IoT-enabled intelligent pedestrian crossing signal light system with violation tracking
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Rupasinghe, R. A. I. M.; Ranasinghe, R. A. J. B.; Moragoda, Y. G. D.; Navodya, W. D. I.; Premasiri, R. H. M. D.; Chethana, E. J. K. S.; Seneviratne, J. A.; Gunawardana, K. D. B. H.
The urban pedestrian crossing environment presents numerous challenges in ensuring the safety of pedestrians and maintaining smooth traffic flow. Traditional pedestrian signaling systems operate on fixed timings and have limited capabilities, making it difficult to manage the complexities of modern urban traffic effectively. This research introduces an innovative system for pedestrian crossing signal lights integrated with violation tracking and real-time data analytics to improve pedestrian safety and smooth traffic flow. This encompasses computer vision for pedestrian detection, machine learning (ML) for predictive analysis, adaptive signal light timers, sirens for violation deterrence, and IoT components for seamless real-time operation. The presented methodology combines real-time pedestrian detection, adaptive signal light timing, weather detection, and IoT integration so that all these subsystems work smoothly. The issues resolved include integrating image processing with hardware, selecting an efficient pedestrian detection model, optimizing camera angles for accurate detection, and transitioning from an Arduino to a Raspberry Pi 4 Model B. The Raspberry Pi offered better processing power, enabling faster and more complex data handling. A case study was done at a location proximate to the University of Kelaniya, and the average crossing time taken for the pedestrian crossing was recorded as 18.5 seconds, which can be factored using databases with larger data sets and simple ML models based on the day of the week. The issues that were resolved include integrating image processing with hardware, selecting an appropriate pedestrian detection model such as a Convolutional Neural Network (CNN) that works well within the outdoor environment, setting optimal camera angles for accurate pedestrian detection, and transitioning from an Arduino to a Raspberry Pi 4 Model B for enhanced processing capabilities. Integrating image processing with hardware posed challenges due to the need for real-time data transmission and processing, which required seamless communication between the software and hardware components. The pedestrian detection model was chosen based on its accuracy, speed, and ability to perform well in varying lighting and weather conditions. The transition to the Raspberry Pi 4 Model B, with its superior processing power and memory compared to the Arduino, allowed the system to handle more complex tasks, such as real-time data analysis and multiple input streams, significantly improving performance and efficiency. A custom dataset of overhead views of pedestrians was created, with images acquired from a similar environment within the university, manually labelling the images and achieving an 80% accuracy after training on Google Collab. The real-time data processing system is vital in making dynamic signal timing changes, tracking violations to encourage safe pedestrian behavior, and managing pedestrian and vehicle traffic flow. These findings endorse the broader adoption of intelligent systems, for innovative city projects toward safer and more efficient urban environments.
Thought identification through visual stimuli presentation from a commercially available EEG device
(Department of Industrial Management, Faculty of Science, University of Kelaniya Sri Lanka, 2021) Gunawardhana, M. P. A. V.; Jayatissa, C. A. N. W. K.; Seneviratne, J. A.
Thought identification has been the ultimate goal of brain-computer interface systems. However, due to the complex nature of brain signals, classification is difficult. But recent developments in deep learning have made the classification of multivariate time series data relatively easy. Studies have been carried out in the recent past to classify thoughts based on signals from medical-grade EEG devices. This study explores the possibility of thought identification using a commercially available EEG device using deep learning techniques. The crucial part of any EEG experiment is contamination-free data collection. Keeping the subject’s mind concentrated only in the decided state is important, yet challenging. To address this issue, we have developed a graphical user interface (GUI) based program that allows stimulus controlling and data recording. With the use of the low-cost commercially available EEG device, accuracies up to 89% were achieved for the classification of high contrast signals. However, tests on complex thought identification did not produce statistically significant results over the chance accuracy.
Use of cuprous oxide thin film semiconductors for dissolved oxygen sensing: A preliminary study
(Faculty of Science, University of Kelaniya, Sri Lanka, 2021) Wijesooriya, H. E.; Seneviratne, J. A.; Jayathilaka, K. M. D. C.; Wijesundera, R. P.
Monitoring and maintaining the quality of water is extremely important as it can severely affect the health of humans as well as animals. Dissolved oxygen is one of the key indications of water quality. Cuprous Oxide (Cu2O) semiconductor material is an ultra-low cost, environmentally friendly, earth abundant material which is considered as a green alternative to many sensing applications. Therefore, Cu2O thin film semiconductors could potentially act as a dissolved oxygen sensor due to their unique electrical features. Generally, a significant change in the electrical conductivity is caused by the adsorption of molecules on the surface of Cu2O semiconductor material. In this investigation, the possibility of sensing dissolved oxygen using Cu2O thin film semiconductors was explored. Cu2O thin films were grown using the electrodeposition technique on titanium (Ti) substrates at -200 mV vs Ag/AgCl for 60 minutes in an electrochemical cell containing an acetate bath of 0.1 M sodium acetate and 0.01 M cupric acetate. The temperature of the bath was maintained at 55 ℃ and stirred at a speed of 50 rev/min. The Ag/AgCl electrode was used as the reference electrode, while the platinum electrode was the counter electrode. Prepared Cu2O thin films were characterized in two-electrode systems using resistance measurements at ambient conditions in two different Deionized (DI) water volumes of 100 ml and 200 ml while aerating oxygen at a constant rate at the room temperature. Significant linear change in resistance was observed with increasing dissolved oxygen concentration under ambient condition in both cases. In comparison with experiment carried out with 100 ml of DI water volume, it was observed that in 200 ml DI water volume it takes more than twice the time to saturate with oxygen. Furthermore, it was observed that the constant resistance value of the system at the saturation of 200 ml DI water volume was higher than the constant resistance obtained at the saturation of 100 ml DI water volume. This preliminary investigation revealed that Cu2O thin films could use to monitor dissolved oxygen. However, further investigations need to be performed to optimize the dissolved oxygen sensing property of Cu2O thin films.
Wireless pager system for enhancing emergency communication in hospital environment
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Gunarathna, T. G. L.; Adhikari, A. M. N. D. S.; Bandara, K. D. Y.; Gunawardana, K. D. B. H.; Seneviratne, J. A.; Perera, M. H. M. T. S.
Maintaining fast and efficient communication between hospital staff is critical to ensure patient safety during emergencies. However, challenges such as the lack of Global System for Mobile Communications (GSM) signals in countries like Sri Lanka and the risk of using cable communication during hazardous weather conditions further complicate emergency communication. This paper proposes a wireless pager system utilizing LoRa (Long Range) technology to facilitate seamless interaction between doctors, nurses, and other supportive and administrative staff in a hospital. LoRa operates on sub-gigahertz frequencies, providing robust signal penetration and extended range, making it ideal for hospital environments where walls and infrastructure often disrupt traditional signals. The proposed system consists of three primary modules: the Ward Module, Central Hub, and Doctor Module. The Ward Module, placed in hospital wards, allows nurses to trigger emergency alerts by selecting an available doctor. It also provides status updates on message delivery and doctors' responses. The Central Hub acts as the system's control center, maintaining a database of doctors and wards, managing doctor availability, registering new entries, and logging communication transactions. It utilizes a web-based application to handle and collect data, which runs on the Central Hub, streamlining data management and access. The Hub also backs up data to the cloud and stores it locally during internet outages, synchronizing once the connection is restored. The Doctor Modules enable doctors to log their presence by selecting their ID from a list obtained from the Central Hub. This login data is updated in the Central Hub and shared with the Ward Modules. Upon receiving an emergency alert, doctors can respond by accepting, canceling, or forwarding the message, with the updated status being communicated back to the Ward Module. The system was tested in a simulated hospital environment using two Ward Modules, two Doctor Modules, and a Central Hub, covering a 200m distance. Both the Ward and Doctor Modules were built using ESP32 microcontrollers with LoRa modules operating at 433 MHz, while the Central Hub was developed using a Raspberry Pi single board computer with a LoRa module. The system demonstrated reliable performance, maintaining stable communication across the test range. It also demonstrated potential for larger hospitals, with extended range possible through proper antenna configuration. A 96% success rate was recorded, with message transmission in under 2 seconds. While LoRa offers robust long-range communication with low power use, its limited bandwidth poses challenges for large data transmission. However, for emergency pager systems, the trade-off between power efficiency and data capacity is acceptable. The system operates independently of traditional communication infrastructure, providing hospitals with a sustainable and resilient solution for emergency communication. It streamlines emergency response in hospital wards by enabling realtime communication and status updates between staff, ensuring fast and accurate transmission of critical information. This enhances the efficiency of interventions and improves patient care outcomes.