http://drr.vau.ac.lk/handle/123456789/238 Mon, 21 Sep 2026 20:46:11 GMT 2026-09-21T20:46:11Z http://drr.vau.ac.lk/handle/123456789/2238 Federated Learning for Detecting Anomalies in IoT-Driven Smart Home Systems Thevarajan, J.; Ravi, V.; Kirushanth;, S.; Ganesalingam, V.; Ahmadon, M. A.; Che Lah, N. S. B. The rapid proliferation of Internet of Things (IoT) devices in smart home environments has amplified significant cybersecurity challenges due to their outdated firmware, limited computational resources, and weak protection mechanisms. Traditional centralized anomaly detection approaches are ineffective in these environments owing to single points of failure, privacy risks, and limited scalability. To address these challenges, this study proposes a novel federated learning framework that integrates a lightweight hybrid architecture combining 1D Convolutional Neural Networks (1D-CNN) with shallow autoencoders to support both supervised and unsupervised anomaly detection. The supervised component enables robust binary classification of known attack patterns, while the unsupervised reconstruction mechanism enhances detection of previously unseen anomalies, thus balancing accuracy with adaptability. Distinct from prior approaches, the proposed framework employs dynamic quantization within the federated learning process, enabling substantial reductions in resource consumption without compromising detection performance. Experiments conducted on the TON-IoT dataset, with SMOTE-based class balancing and PCA-driven dimensionality reduction, under a realistic non-IID distribution across three federated clients, demonstrate that the proposed model achieves 97.39% test accuracy. This model is designed for deployment on resource-constrained devices, employing post-training dynamic range quantization that reduces the model size by 81.5% (42.60 KB to 7.87 KB) without compromising accuracy. The experiment performed using Raspberry Pi 4 confirmed sub-millisecond inference and real-time responsiveness, validating the practical viability of the framework for efficient anomaly detection in smart IoT systems. Wed, 01 Jan 2025 00:00:00 GMT http://drr.vau.ac.lk/handle/123456789/2238 2025-01-01T00:00:00Z http://drr.vau.ac.lk/handle/123456789/2237 Design and Development of Weigh-In-Motion Using Vehicular Telematics Kirushanth, S.; Kabaso, B. Identifying overloaded vehicles on a highway is essential for the safety of vehicles on the road as well as for the performance monitoring of highway infrastructure and planning. Traffic enforcement uses various weigh-in-motion (WIM) methods. Since Vehicular Telematics (VT) is favoured in the transport industry, using it for building a new WIM system to infer the payload of a vehicle at any road segment would be beneficial for the transport industry. This paper presents the effort taken to use VT data from onboard diagnostics modules and smartphones to infer the payload of a vehicle. The experiment done to find the correlation between VT data and the payload of a vehicle is discussed. Feature engineering was done; nine different settings were tested to find the best regression model. A multiple nonlinear regression model produced significant a p value of 6.322e-08 and an R-squared value of 0.8736. Results support the notion of using the VT data for nonintrusive measurement of the weight of a vehicle in motion. Wed, 01 Jan 2020 00:00:00 GMT http://drr.vau.ac.lk/handle/123456789/2237 2020-01-01T00:00:00Z http://drr.vau.ac.lk/handle/123456789/2224 Stability Analysis of Sri Lankan Tea Export Markets Using Markov Chain Approach Kayathiri, T.; Laheetharan, A. Among plantation crops, tea has been playing a significant role in the agrarian economy as it adds substantially to the GDP of thc country. Stability analysis of tea export market by the economic community in any country serves several purposes viz business forecast for wise decisions; the assessment of the growth of the economy; and, the future design of tea plantations. Contribution of the tea production within the agriculture sector leads a dominant role in providing employment opportunities in Sri Lankan economy. "lhe aggregate data of tea export from Sri Lanka to various countries is collected from the Sri Lanka Tea Board and the Central bank reports for the period covers from 2009 to 2017. The objective ofthe study is to analyze the dynamic changes oftea export volume from Sri Lanka to tea markets in different countries employing the Markov chain model. The aggregate data from Markov chain models explain that the present observed proportion of one country is stochastically related to recent past observed proportion of the countries including itself. The transition probabilities are estimated using the non-linear programming technique under the minimization of mean absolute deviation method. In addition, the steady state probabilities and mean recurrence times have been obtained. The results have shown Syria and Iraq as the stable destinations for Sri Lankan tea exports. Ihe other traditional importing countries such as Russia, Turkey and Iran have recorded low retention probabilities, indicating unstable export share to these markets. The long run stable probabilities revealed that Sri Lanka retains the trade partnership with traditional importing countries except Syria. Tue, 01 Jan 2019 00:00:00 GMT http://drr.vau.ac.lk/handle/123456789/2224 2019-01-01T00:00:00Z http://drr.vau.ac.lk/handle/123456789/2160 Revan topological indices of supramolecular Fushine acid useful in medical applications Gunawardhana, D.C.; Sino, A. M. F. S.; Moulis, K.; Perera, K. K. K. R.; Lanel, G. J. Chemical graph theory plays a pivotal role in mathematical chemistry by representing chemical structures as graphs, with vertices denoting atoms and edges denoting chemical bonds. Topological indices, numerical invariants derived from such graphs, have been widely employed in quantitative structure property relationship (𝑄𝑆𝑃𝑅) and quantitative structure activity relationship (𝑄𝑆𝐴𝑅) studies. These indices correlate molecular structure with physicochemical and biological properties and have become crucial tools in drug design. Supramolecular chemistry, which studies entities formed bymolecular self-assembly through non-covalent interactions, offers an exciting avenue for designing complex molecular architectures. In this work, we investigate the supramolecular structure of Fuchsine (C₂₀H₁₉N₃HCl), a magenta dye of significant microbiological and histological importance. We construct a supramolecular sheet, denoted [𝑚, 𝑛], comprising 𝑚 × 𝑛 units of Fuchsine molecules. The corresponding chemical graph is simple, connected, and finite, consisting of 38𝑚𝑛 + 𝑚 + 𝑛 vertices and 42𝑚𝑛 edges, which are further classified by the Revan degrees of their end vertices. We derive closed-form expressions for several Revan degree-based topological indices of the supramolecular Fuchsine sheet, including the first and second Revan indices(𝑅1 and 𝑅2), Atomic Bond Connectivity Revan index(ABCR), Geometric-Arithmetic Revan index(𝐺𝐴𝑅), the first and second hyper Revan indices(𝐻𝑅1 and 𝐻𝑅2), the first and second modified Revan indices(𝑚𝑅1 and 𝑚𝑅2) , forgotten Revan index(𝐹𝑅). A detailed numerical and graphical analysis demonstrates that all these indices increase monotonically with the parameters m and n, reflecting the scaling behaviour of the supramolecular structure. Among the indices studied, the first hyper Reven index exhibits the highest values, whereas the first modified Revan index yields the lowest. Our findings provide a comprehensive mathematical characterization of the supramolecular Fuchsine graph, offering valuable insights for modelling and predicting the properties of complex chemical systems using topological descriptors. Wed, 01 Jan 2025 00:00:00 GMT http://drr.vau.ac.lk/handle/123456789/2160 2025-01-01T00:00:00Z