Department of Computer Science
Permanent URI for this communityhttps://repository.cuilahore.edu.pk/handle/123456789/16
Browse
8 results
Search Results
Item Balancing Fertilizer Usage with Real-Time NPK Data: A Precision Agriculture Model for Sustainable Agriculture in Pakistan(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Jawad Hassan; CIIT/SP24-RCS-005/LHR; Dr. Tariq Umer; LHR TO 10036Pakistan is facing a lot of problems in the agriculture sector as the use of imbalanced fertilizers and inefficiency of nutrients management is going to cause the very low yield of the agricultural fields. It is a matter of fact that traditional agricultural methods often lead to the inappropriate distribution of fertilizers that affects the crop yields, hikes soil infertility, and devastates the environment even more. Either more or less fertilizers usage disrupts the nutrient balance in the soil thereby creating longer agricultural inefficiencies. Utilizing real-time soil and climate monitoring and analysis can provide the technological power to deal with these challenges as AI-driven precision agriculture gains popularity. This research is dedicated to the development of an intelligent fertilizer recommendation system by using sensor-based NPK analysis that is combined with Al decision-making. This system will use prompt and accurate data from soil nutrient sensors, weather conditions, and crop requirements to provide recommendations of a proper fertilization dosage to specific fields. The novel approach seeks to minimize fertilizer wastage, to lower the environmental impact, and to maximize crop yields. The new method aims to reduce the waste of fertilizer, to keep the environmental impact at a low level, and to make the harvest bigger. One of the approaches involves the use of IoT-based soil monitoring systems to gather real-time data, training of machine learning models on historical soil and crop data for nutrient requirements prediction, and the development of a user-friendly decision-support system for farmers. This will be achieved through the provision of actional insights for the farmers who will later take the necessary course of action in their fertilizer applications. Optimizing nutrient management through the use of the new system is the way to go to stop the degradation Environmental issues like soil acidification andItem AI-Driven Resource Allocation in Edge Metaverse for the Internet of Medical Things (IoMT)(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Saira Shairi; CIIT/SP24-RCS-023/LHR; Dr. Tariq Umer ; LHR TP 10042Healthcare delivery is increasingly dependent on the Internet of Medical Things (IoMT), yet the rapid growth of heterogeneous devices and dynamic network conditions make resource allocation, latency control, and privacy preservation critical challenges. In modern healthcare, ensuring timely and accurate activity recognition is essential for improving patient care and enabling real-time clinical decision-making. The variability in device capacity and communication links restricts the effectiveness of conventional resource management strategies. This research presents a predictive, privacy-preserving framework for intelligent resource optimization in IoMT environments, leveraging Federated Reinforcement Learning (FRL) with a Deep Q-Network (DQN) to support distributed decision-making without exposing sensitive health data. The study integrates a Hybrid FRL-ConvLSTM model for activity recognition, enabling accurate workload prediction to guide scheduling and offloading strategies between device and edge layers. Simulation experiments under diverse workloads and network conditions confirm that the proposed approach achieves significant improvements in reducing delay, and bandwidth efficiency. The proposed model is effective for real-time healthcare monitoring and demonstrates that combining predictive deep learning with FRL-based optimization can significantly enhance reliability, and privacy in IoMT systems, it gives accuracy of 99.5% .Item Multi-Pollutant based Hybrid Framework for Green Smart Cities using Machine Learning Techniques.(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Usama Masood; FA21-RCS-013; LHR TP 8676; Dr. Tariq UmerAchieving a sustainable environment is one of the most emerging issues discussed in the smart cities concept. Due to the rapid population growth in the world, the concentration of greenhouse gases is increasing day by day. A lot of research studies have focused on different techniques and technologies to reduce environmental pollution. To achieve a sustainable environment, it is important to consider the multi-pollutant factors involve in polluting the environment. In MS thesis, we proposed a multi-pollutants-based intelligent hybrid framework using machine learning that considers the multiple sources of pollution in the city environments. The research emphasizes a comprehensive comparisons of machine learning algorithm for predicting the air quality index in smart cities. In this research. The framework considers the concentration of the pollutants in the environment and will make intelligent predictions on their combined effect on the environment as well as the individual groups based on the similar characteristics of gases concerning the guidelines of WHO. The effect of the concentration of multi-pollutants on air quality and water quality will be analyzed. Machine Learning Techniques including Linear Regression, Support Vector Regression, Random Forest Regression, Decision Tree Regression, and Long Short Term Memory (LSTM) will be applied to predict air quality index and for the classification of pollutants. The evaluation measure used in this research is Accuracy, Precision, F1 score and Recall to find the accuracy of these models.Item Design and Development of a Secure and Light Weight User Authentication Scheme for Smart Home(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2022) Shiza Gul; SP21-RCS-011; LHR TP 8060; Dr. Tariq UmerThe Internet of Things (IoT) is becoming more advanced day by day, in which smart homes provide different services of home to people. Users like the comfort of the use of home service that smart devices provide. But here is privacy and security in smart homes because smart devices and users perform communication across insecure channels. As smart devices with limited resources do not have the computing and storage capacity to create and execute existing methods that often require complicated computations, Smart Home Area Network (HAN) system experiences major problems in establishing strong authentication methods. Furthermore, the diverse nature of IoT nodes makes it difficult to create end-to-end device authentication in IoT systems. IoT devices commonly use poor authentication protocols. The primary barrier against numerous security attacks on IoT networks is authentication. To strengthen IoT privacy and security against frequent attacks like impersonation and replay attacks, this thesis analyses the research gaps and provides a new security mechanism and authentication technique. By providing a lightweight mutual-authentication and key exchange protocol for IoT based identity and continuously chained hash in HAN, this research improves IoT privacy and security. Using temporary symmetric keys and identities, devices may mutually authenticate and create a connection with the gateway node in an untraceable method. To verify the sender's identity, the Continuous chained-hash approach is presented (through a response challenge). The session keys are often updated after each data transmission to maintain forward secrecy. The proposed method provides authentication while ensuring that IoT device transactions cannot be tracked. Three different techniques are used to assess and simulate the security of the protocols; formal analysis that used the Burrows-Abadi-Needham logic (BAN), informal analysis, and model-checking that used the automated validation of Internet security protocols and applications (AVISPA) tool. The cost and effectiveness of the proposed method are evaluated and compared with other relevant protocols. The outcomes demonstrated that our techniques are more effective.Item A Digital Twin based Approach for Healthier Smart City Environment(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Khazina Naveed; FA21-RCS-001; Dr. Tariq UmerAir quality, toxic substances, climate change, pollution, and occupational hazards are the environmental factors that have a great impact on human and environmental health. The presence of pollutants and contaminants in the environment can have detrimental effects on human health, including respiratory issues, cardiovascular ailments, cancer, neurological disorders, and various other illnesses. Certain groups, such as children, older adults, and individuals with pre-existing health conditions, are more vulnerable to the health impacts of environmental hazards. Air Quality Index is used to depict the air quality of an area. This thesis delves into studying the utilization of Digital Twin models as an innovative strategy for creating a sustainable smart city environment by accurately forecasting Air Quality Index (AQI) using time series analysis. The study applies different deep learning models, namely Gated Recurrent Units (GRU), Long Short Term Memory (LSTM), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and deep Artificial neural networks (ANN) for precisely predicting the AQI levels for a healthier environment. The time-series historical data of Delhi city has been gathered from the year 2015 to 2020 and is preprocessed before proceeding to train and validate different deep learning models. The six selected machine learning algorithms have been implemented and it has been observed that CNN offers unparalleled accuracy compared to other evaluated models making it highly effective for precise forecasting. The CNN-1D-2 layer yielded the best results with root mean squared error (RMSE) reaching 3.010343, the mean absolute error (MAE) reaching 1.706329, the mean absolute percentage error (MAPE) reaching 0.013216, and the R2 reaching 0.99931. The digital twin model is developed by incorporating InfluxDB and Grafana. InfluxDB is an open-source platform that has been used to store the historical and real-time data of AQI using python. The Grafana online platform is utilized for data visualization and management and facilitates real-time monitoring. The 3D model of the city is developed in Blender and then the 3D file is exported to Microsoft Azure Digital Twin explorer to develop a digital twin using Digital Twin Definition Language (DTDL). The digital twin model displays the forecasted value of AQI and different pollutants along with their previous trends in the form of graphs. The 3D model of the city is covered with different colors based on ranges or thresholds defined for AQI. The findings contribute to the advancement of smart city development by utilizing digital twin-based approaches for creating healthier and more sustainable urban environmentsItem A Digital Twin Assisted Disease Detection for Leafy Green Vegetables Using Federated Learning in Smart Greenhouse(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Iqra Pervez; CIIT/SP23-RCS-016/LHR; Dr. Tariq Umer; LHR TP 9703Agricultural production is seriously vulnerable to the rapid evolution of plant diseases in greenhouse environment. In modern agriculture, ensuring timely and accurate disease detection in leafy green vegetables is crucial for improving crop yield and sustainability. The effects of environmental temperature, humidity on plant’s health which restricts the efficacy of current disease management techniques in greenhouses. This research presents a Digital Twin-assisted model for disease detection in lettuce plants, leveraging Federated-based Learning to enable privacy-preserving and distributed model training within a smart greenhouse environment. The study integrates five pre-trained models AlexNet, MobileNet, ResNet50, EfficientNetB3, and VGG16 to evaluate their effectiveness in classifying plant diseases. This study utilise a lettuce diseases dataset from Kaggle. Among the tested models, VGG16 achieved the highest accuracy of 98%, demonstrating its superior capability for disease classification in resource-constrained settings. Additionally, a correlation analysis between humidity, temperature and disease occurrence was conducted using Unity 3D and Microsoft Azure, providing deeper insights into the environmental conditions influencing plant health. The proposed model is useful for monitoring crops in real time and shows that combining Digital Twin technology with Federated-based Learning can improve decision-making in smart farming systems.Item Neural Network-Based Anomaly Detection in Cryptocurrency Transactions(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Rashida Noreen; CIIT/SP23-RCS-025/LHR; Dr. Tariq Umer; LHR TP 9705Cryptocurrency, especially the rapid growth in adopting bitcoins has led to a significant increase in fraud and illegal transactions, which faces serious challenges for financial integrity and digital trust. Methods of detecting traditional fraud are often reduced to detecting sophisticated patterns within the blockchain network due to decentralized and dynamic nature of transaction flows. This thesis presents a graph-based deep learning approach to detect discrepancy in cryptocurrency transactions, which uses a large, label dataset of bitcoin transactions collected over time. The functioning begins with the collection and prepressing of the elliptical dataset, which includes more than 230,000 edges representing more than 200,000 transactions and money flows represented as nodes. Dataset consists of three classes: license, illegal and unknown transactions. To extract meaningful insight from this data, a broad time-based and network-based feature engineering was performed. After feature engineering, the dataset was converted into a graph structure suitable for processing by the graph neural network (GNNS). Four state -of -the -art models trained and evaluated: Graph Attention Network (GAT), GATV2, Siamese GNN, Relational Graph Convolution Network (RGCN), and graph transformers. The model was trained and validated using stratified partitions to ensure balanced class representation. In the models, the Siamese GNN achieved the highest performance with an accuracy of 95.67% GAT followed closely with 94.90% accuracy, while graph transformers 94.60% and GATV2 achieved 94.46% competitive but slightly less results. RGCN showed 93.77% the least effective performance, especially struggling with unknown transactions classification. The results strongly suggest that graphs can play an important role in increasing the accuracy and reliability of the system that detects fraud in the blockchain environment, especially in accordance with dynamic and asymmetrical graphs.Item Design and Development of Trust-Based Secure Access Control in Metaverse for Wearable Devices(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Zonash Faysal; CIIT/SP23-RCS-030/LHR; Dr. Tariq Umer; LHR TP 9707The sudden emergence of the metaverse has expanded options for immersive digital environments, especially through head-worn wearables that function as portals to virtual spaces. But then the integration of wearables with the metaverse has some serious security, privacy, and trust implications. In this dissertation, we propose a trust- based secure access control mechanism dedicated to wearable devices inside the metaverse. The proposed approach makes sure the trustworthiness of the network by examining four trust parameters, i.e., security, privacy, integrity, and availability. The trustworthiness measurement of the system is described, and a weighted trust score is computed to measure the reliability of the system, and a comprehensive authentication protocol is designed, combining multi-factor authentication principles, encryption, certain hashing mechanisms and a session key generation process. Using the AVISPA tool, the architecture is validated and discusses the resilience to the well known attacks impersonation, replay and theft. The results give a story to the efficacy of the lightweight framework to establish a secure environment for wearables in the metaverse. This research adds to the growing bank of knowledge in the field of improving the safety, security and reliability of virtual environments for end users.