Research Publications
Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/4194
This main community comprises five sub-communities, each representing the academic contribution made by SLIIT-affiliated personnel.
Browse
1106 results
Search Results
Item Unknown Securing Distributed Disaster Early Warning Networks(Sri Lanka Institute of Information Technology, 2026-05-21) Senarathne, K; Dias, D; Prasanna, RThis paper proposes a secure and cost-effective communication and management approach for distributed disaster early warning networks to ensure reliable, low-latency alert dissemination. While decentralized architectures reduce latency by enabling direct device communication, they are highly vulnerable to attacks due to their exposure in public community settings and the challenges posed by Network Address Translation (NAT) and firewalls. Traditional NAT traversal techniques often lack the built-in security required for such sensitive infrastructure. We identify Nebula, an open-source overlay networking tool, as a viable solution because it is lightweight enough for low-resource devices such as the Raspberry Pi typically found in decentralized early warning systems, supports NAT traversal, and provides endto- end encryption. Preliminary experiments conducted in an AWS environment—designed to simulate real-world Raspberry Shake sensor deployments—confirm Nebula’s ability to establish direct peer-to-peer (P2P) connectivity with a single hop and effectively encrypt trafficItem Unknown From Beach to Cliff: Adapting CoastSnap Citizen Science for Coastal Cliff Change Detection Using ML-Assisted Image Registration and Prompted Segmentation(Sri Lanka Institute of Information Technology, 2026-05-21) aramillo-Velez, A; Gamlath, S; Chandirakumar, M; Prasanna, R; de Vilder,S; McColl, S; Tan, M.L; Stewart, C; Ambegoda, T.DCoastSnap is a citizen science tool that uses repeat smartphone photographs from fixed stations to monitor coastal change, yet it has rarely been applied to coastal cliffs. We test a workflow for cliff change screening using a controlled pilot CoastSnap station at Ōnaero, New Zealand (iPhone 12), integrating (i) Machine Learningassisted ground control point (GCP) transfer for image registration, (ii) pinhole camera calibration and reprojection-based geo-rectification onto a curved cliff-surface model, and (iii) prompted segmentation for isolating cliff-related features. Auto-GCP benchmarking shows a tradeoff between precision and robustness: the Scale-Invariant Feature Transform (SIFT) model achieves low localisation error when successful, but is sensitive to changes in illumination. In contrast, the Local Feature ransformer (LoFTR) model provides a higher detection yield and more stable performance, but is sensitive to the threshold used. Nevertheless, it is well suited to operational use with human-inthe- loop verification. Calibration produced consistent parameters across repeat images, with focal length estimates matching device specifications. Visual segmentation based only in Regions of Interest (RoI) often merged adjacent objects, while text-guided Segment Anything Model (LangSAM) delineated cliff and debris more reliably than fracture-like features. Preliminary detection of the supply and removal of debris highlights the potential for quantifying the frequency and magnitude of mass movements at each cliff.Item Unknown A Dynamic LUCIS Framework for Identifying Urban-Hazard Conflicts in the Kelani River Basin Using Near Real-Time Earth Observation Data(Sri Lanka Institute of Information Technology, 2026-05-21) Aloka, O; Nayanajith, B; Dassanayake, SRapid urbanisation in Sri Lanka frequently bypasses formal planning, resulting in concealed land-use conflicts where infrastructure encroaches on disaster-prone areas. Traditional Land-Use Conflict Identification Strategy (LUCIS) models are based on static, outdated survey data and fail to capture realtime risks. This temporal lag leads to undetected unauthorized settlements in hazard zones increasing disaster vulnerability. This study introduces a Dynamic LUCIS framework utilizing near real-time Earth Observation (EO) data from Google Earth Engine, specifically the Dynamic World and Open Buildings datasets. Focused on the Kelani River Basin, the model identifies critical risk conflicts between urban development and flood/landslide safety zones. Validation using high-resolution drone imagery yielded an overall accuracy of 82.84%, F1-score of 0.83 and a Kappa coefficient of 0.66. The results demonstrate that EO-driven models can identify unauthorized settlements in hazard-prone areas with high precision, providing a scalable tool for resilient urban governance.Item Open Access Voice Communication over Meshtastic Networks(Sri Lanka Institute of Information Technology, 2026-05-21) Uduwaka, S; Gunawardana, W M T V; Lakshan, W D T; Pathirana, R P S; Prasanna,R; Dias, D; Gayan, SVoice communication is essential in post-disaster scenarios where users may be unable to type due to injury, stress, or environmental constraints. While Meshtastic provides a decentralized LoRa-based mesh network for long-range messaging, it does not natively support voice and is limited by low data rates and strict regulatory duty-cycle requirements. This paper investigates two methods for enabling voice over such networks: waveform-based audio packetization and a semantic speech-to-text (STT) / text-to-speech (TTS) approach. Preliminary experimental evaluations demonstrate that waveform-based methods, even when using low-bitrate codecs like Codec2, are impractical for urgent communication; delivering a 3-second voice message requires hundreds of packets and several minutes due to ransmission latency and duty-cycle constraints. In contrast, the proposed offline STT/TTS method transcribes speech locally, transmits it as a standard text payload, typically in a single packet and reconstructs the audio at the receiver. This semantic approach reduces delivery time from minutes to seconds, while maintaining high reliability and regulatory compliance. The study concludes that semantic communication is the most viable solution for voice interaction in bandwidth-constrained LoRa mesh environments.Item Open Access Development of a MEMS-based Earthquake Dataset using the Raspberry Shake Network in New Zealand(Sri Lanka Institute of Information Technology, 2026-05-21) Samiha, T.Z; Ravishan, DTraditional seismic monitoring is often limited by the high cost of instrumentation and logistical barriers, hindering the expansion of earthquake monitoring networks especially in under-resourced regions. Low-cost MEMS-based sensors offer a scalable alternative, but require specialized datasets to train machine learning models adapted to their unique noise characteristics and sensitivity profiles. To address this, we systematically collected waveforms from approximately 4,000 earthquakes (magnitude 2.7 to the highest recorded) recorded across 89 Raspberry Shake stations in New Zealand from 2020–2025. Events were matched to nearby stations based on epicentral distance criteria (100 km for M 2.7–5.5, 150 km for M>5.5). A staged filtering pipeline using PhaseNet, EQTransformer, and GPD models, cross-validated with theoretical TauP arrivals, was applied to ensure phase pick quality across three confidence tiers. The final curated dataset comprises 918 high-confidence waveforms with validated P and S wave arrivals, alongside approximately 16,035 total waveform records spanning all quality tiers. This dataset addresses the critical scarcity of labeled training data for low-cost seismic instrumentation, enabling the development of phase pickers specifically calibrated for MEMS sensors.Item Open Access Large Language Model Pipelines for Crisis Intelligence: Automated Help Request Classifier(Sri Lanka Institute of Information Technology, 2026-05-21) De Silva, L.D.R.E.This research examines the development and deployment of a Crisis Intelligence Pipeline, engineered to address the catastrophic information surge following Cyclonic Storm Ditwah in Sri Lanka [1]. The system integrates advanced prompt engineering methodologies, specifically few-shot learning, Chain-of-Thought [2], and Tree-of-Thoughts [3] to transform unstructured humanitarian data into actionable logistics strategies. Through rigorous stability testing under varying temperature gradients, the study identifies a "Safe Mode" for deterministic triage and a "Chaos Mode" that highlights the risks of model hallucination in high-stakes environments. Furthermore, the pipeline utilizes Pydantic-based schema validation [4] and token economics [5] to ensure scalability and cost-efficiency. Results indicate that structured reasoning architectures significantly enhance the precision of resource allocation and signal-to-noise discrimination, providing a vital technological framework.Item Open Access Smart-Camp Box: An Integrated IoT and Artificial Intelligence Framework for Safety, Communication, Learning Assistance, and Resource Optimization in Disaster Relief Camps(Sri Lanka Institute of Information Technology, 2026-05-21) Kashmira S S; Hansana G P; Deerasinghe A D S N S; Jayakody, A; Lokuliyana, SNatural disasters continue to inflict devastating consequences on communities across South Asia, with Sri Lanka ranking among the most frequently affected nations in the region. When disasters strike, temporary relief camps serve as critical shelters for displaced populations; however, existing systems fail to address three persistent operational challenges simultaneously: camp-level flood and landslide prediction, psychological and attentional readiness assessment for displaced children, and resilient communication under degraded network conditions. This paper presents the Smart Camp Box, an integrated, portable IoT-based framework that addresses all three dimensions through tightly coupled sub-systems. The first sub-system provides location-aware environmental risk monitoring using GPS/GNSS, IoT sensors, DEM integration, and a machine learning prediction model. The second introduces the Psycho-Attentional Gated Educational System (PAGES), an offline-first application for assessing and supporting displaced children's cognitive readiness. The third proposes a Semantic-Aware Adaptive Message Prioritization (SAAMP) framework for reliable MQTT communication under constrained network environments. Currently in active development, the Smart Camp Box represents a paradigm shift toward proacItem Open Access Space-Ground Hybrid AI System for Flood and Landslide Monitoring Using Multi-Level Data Fusion(Sri Lanka Institute of Information Technology, 2026-05-21) Withanachchi, W U; Ranasinghe, M T EFloods and landslides represent critical hydrometeorological hazards whose frequency and intensity are amplified by extreme precipitation, deforestation, unplanned urban expansion, and inadequate land-use management. Conventional Early Warning Systems (EWS) are typically constrained by reliance on single-source data streams such as satellite observations, IoT sensor networks, or community reports, resulting in fragmented situational awareness, elevated predictive uncertainty, and limited lead time. To address these limitations, this study proposes a Space-Ground Hybrid AI framework that integrates heterogeneous data sources through a hierarchical multi-level fusion architecture. The proposed system combines convolutional neural networks (CNNs) for spatial feature extraction, long short-term memory (LSTM) networks for temporal sequence modeling, and gradient-boosted ensembles for hazard-specific prediction, enabling comprehensive modeling of both large-scale environmental patterns and localized dynamic processes. These outputs are systematically fused at feature, decision, and risk levels to derive a Unified Hazard Risk Index (UHRI), which provides an interpretable and operationally actionable representation of multi-hazard risk. Experimental evaluation using multi-spectral satellite imagery, high-frequency IoT sensor measurements, and simulated community intelligence demonstrates that the proposed hybrid model achieves superior performance, with an accuracy of 91%, an F1-score of 0.89, and a substantial reduction in false alarm rates compared to singlesource baselines. Furthermore, the framework improves warning lead time and maintains robust performance under partial data degradation scenarios, highlighting its resilience and suitability for real-world deployment. Overall, the results validate the effectiveness of multi-modal data fusion and hybrid AI modeling in enhancing predictive reliability, spatial coverage, and temporal responsiveness for disaster early warning systems, providing a scalable foundation for next-generation risk monitoring and management solutions.Item Open Access Beyond Internet Dependency: LoRa for Post- Earthquake Structural Monitoring(Sri Lanka Institute of Information Technology, 2026-05-21) Rallapudi, M.O; Vautier, A; Ngan, B; Perera, P; Prasanna, RAcquiring structural response data during and after earthquakes is essential for assessing threats to critical infrastructure; however, conventional transmission systems rely on Internet connectivity that is often disrupted during major seismic events. This study evaluates Long Range (LoRa) radio as a low-power, infrastructure-independent alternative for transmitting building instrumentation data when conventional networks fail. Field experiments were conducted across three New Zealand cities, testing performance under varying radio configurations, elevations, and urban topographies. Results demonstrate reliable urban communication ranges of up to 1.6 km under optimal settings. Elevation was the dominant deployment factor, with packet delivery ratios (PDR) improving from 0–40% at ground level to near-100% at 500 m radial distance when transmitters were positioned on upper floors. Building-to-building links achieved 75–100% PDR with received signal strength values between −55 and −110 dBm, while terrain features such as hills and dense vegetation caused significant attenuation and occasional link failure at shorter distances. These findings confirm the feasibility of LoRa for post-earthquake structural monitoring and provide quantitative guidance for elevation-aware, terrain-conscious deployment in urban environments.Item Open Access Investigating the Locational Accuracy of Crowdsourced Data in the Context of NSW Wildfires Using NLP and Geocoding Techniques(Sri Lanka Institute of Information Technology, 2026-05-21) Vishwajith, D; Koswatte, SWildfires are major natural disasters, and timely monitoring is essential for effective response. Recent progress in Crowdsourced Data (CSD) and Remote Sensing (RS) has improved wildfire monitoring, however their combined use for spatial validation is still limited. This study examined the locational accuracy of Twitter-based wildfire reports during the 2019-2020 New South Wales (NSW) - Australia wildfire season by comparing them with satellite-derived hotspots. Wildfirerelated tweets were processed using Natural Language Processing (NLP), Named Entity Recognition (NER), and geocoding through the Nominatim API, while hotspot data was obtained from the VIIRS sensor via Digital Earth Australia. Advanced data cleaning and keyword filtering methods were applied to extract relevant geolocated tweets from a large raw dataset, and geocoded locations were rigorously clipped to the NSW boundary to improve spatial relevance. A 5 km grid overlay and statistical tests, including chi-square, binomial tests and Poisson-Based tests, were used to measure the spatial relationship between the two datasets. The results showed a strong positive association, indicating that Twitter reports often align with satellite-validated hotspots, particularly in densely reported areas. Although CSD can be uneven in coverage and limited by geocoding accuracy, a combined approach using NLP, spatial filtering, and statistical validation improves its reliability. The study highlighted the value of integrating social media data with RS, proposed a reproducible framework for spatial accuracy assessment, and provided practical guidance relevant to real-time wildfire monitoring and disaster management systems.