Browsing by Author "Goyal, Nitin"

Now showing 1 - 10 of 10
  • No Thumbnail Available
    Item
    Deep Learning for Monitoring and Managing Oil Spill in Underwater Sensor Networks: Enabling Technologies
    (2026) Goyal, Nitin
    Environmental problems can significantly impact people’s lives and must be addressed. The oil spill is one of the biggest applications of underwater wireless sensor networks (UWSNs) that threaten aquatic life, so it is essential to focus on it. Oil spill count has risen in recent years because of the rise in shipping and marine transportation industries. Timely and accurate detection of oil spills can improve the response process and get the necessary resources to the affected areas more efficiently. This work was motivated by the fact that oil spill detection is critical to maritime protection. Since oil spill detection is a complex process, it faces various challenges. One of the biggest challenges is similar visual appearance which makes it difficult to identify an oil spill and a similar oil slick in synthetic aperture radar (SAR) imagery. Various radar images are frequently utilized for this purpose. Despite being widely used for earth observation, SAR images have noisy and illegible image quality, which makes classification challenging. Previously, detecting and classifying SAR images required manual involvement, which made the process time consuming. Therefore, researchers focused on automating such tasks by incorporating deep learning (DL) techniques. Numerous papers proposed the applications of DL in various radar images for oil spill monitoring but faced multiple problems. This study aims to thoroughly investigate oil spills and their detection techniques, utilizing DL techniques applied to various radar images. These studies have originated in diverse nations with distinct environments. However, compared to other types of radar images, synthetic aperture radar (SAR) images are more effective in pinpointing the location of oil spills. However, they are also rather complex.
  • Thumbnail Image
    Item
    E-Learning environment based intelligent profiling system for enhancing user adaptation
    (Electronics, 2022) Kaur, Ramneet; Gupta, Deepali; Madhukar, Mani; Singh, Aman; Abdelhaq, Maha; Alsaqour, Raed; Breñosa, Jose; Goyal, Nitin
    Online learning systems have expanded significantly over the last couple of years. Massive Open Online Courses (MOOCs) have become a major trend on the internet. During the COVID-19 pandemic, the count of learner enrolment has increased in various MOOC platforms like Coursera, Udemy, Swayam, Udacity, FutureLearn, NPTEL, Khan Academy, EdX, SWAYAM, etc. These platforms offer multiple courses, and it is difficult for online learners to choose a suitable course as per their requirements. In order to improve this e-learning education environment and to reduce the drop-out ratio, online learners will need a system in which all the platform’s offered courses are compared and recommended, according to the needs of the learner. So, there is a need to create a learner’s profile to analyze so many platforms in order to fulfill the educational needs of the learners. To develop a profile of a learner or user, three input parameters are considered: personal details, educational details, and knowledge level. Along with these parameters, learners can also create their user profiles by uploading their CVs or LinkedIn. In this paper, the major innovation is to implement a user interface-based intelligent profiling system for enhancing user adaptation in which feedback will be received from a user and courses will be recommended according to user/learners’ preferences.
  • Thumbnail Image
    Item
    Energy-aware live VM migration using ballooning in cloud data center
    (Electronics, 2022) Gupta, Neha; GuptaK, Kamali; Qahtani, Abdulrahman M.; Gupta, Deepali; Alharithi, Fahd S.; Singh, Aman; Goyal, Nitin
    The demand for digitization has inspired organizations to move towards cloud computing, which has increased the challenge for cloud service providers to provide quality service. One of the challenges is energy consumption, which can shoot up the cost of using computing resources and has raised the carbon footprint in the atmosphere; therefore, it is an issue that it is imperative to address. Virtualization, bin-packing, and live VM migration techniques are the key resolvers that have been found to be efficacious in presenting sound solutions. Thus, in this paper, a new live VM migration algorithm, live migration with efficient ballooning (LMEB), is proposed; LMEB focuses on decreasing the size of the data that need to be shifted from the source to the destination server so that the total energy consumption of migration can be reduced. A simulation was performed with a specific configuration of virtual machines and servers, and the results proved that the proposed algorithm could trim down energy usage by 18%, migration time by 20%, and downtime by 20% in comparison with the existing approach of live migration with ballooning (LMB).
  • No Thumbnail Available
    Item
    FSL-TM: Review onthe Integration of Federated Split Learning with TinyML in the Internet of Vehicles
    (2026) Goyal, Nitin
    The Internet of Vehicles, or IoV, is expected to lessen pollution, ease traffic, and increase road safety. IoVentities’ interconnectedness, however, raises the possibility of cyberattacks, which can have detrimental effects. IoV systems typically send massive volumes of raw data to central servers, which may raise privacy issues. Additionally, model training on IoV devices with limited resources normally leads to slower training times and reduced service quality. We discuss a privacy-preserving Federated Split Learning with Tiny Machine Learning (TinyML) approach, which operates on IoV edge devices without sharing sensitive raw data. Specifically, we focus on integrating split learning (SL) with federated learning (FL) and TinyML models. FL is a decentralised machine learning (ML) technique that enables numerous edge devices to train a standard model while retaining data locally collectively. The article intends to thoroughly discuss the architecture and challenges associated with the increasing prevalence of SL in the IoV domain, coupled with FL and TinyML. Theapproachstarts with the IoV learning framework, which includes edge computing,FL,SL,andTinyML,andthenproceedstodiscusshowthesetechnologiesmightbeintegrated.Weelucidate thecomprehensiveoperationalprinciplesofFederatedandsplitlearningbyexaminingandaddressingmanychallenges. We subsequently examine the integration of SL with FL and various applications of TinyML. Finally, exploring the potential integration of FL and SL with TinyML in the IoV domain is referred to as FSL-TM. It is a superior method for preserving privacy as it conducts model training on individual devices or edge nodes, thereby obviating the necessity for centralised data aggregation, which presents considerable privacy threats. The insights provided aim to help both researchers and practitioners understand the complicated terrain of FL and SL, hence facilitating advancement in this swiftly progressing domain.
  • No Thumbnail Available
    Item
    Joint trust‑based detection and signature‑based authentication technique for secure localization in underwater wireless sensor network
    (2025) Goyal, Nitin
    Ensuring secure and accurate node localization in Underwater Wireless Sensor Networks (UWSN) is a significant challenge, as conventional methods tend to neglect the security risks associated with malicious node interference. This research addresses this issue by proposing a Joint Trust-based Detection and Signature-based Authentication (JTDSA) technique, aimed at improving both the security and accuracy of node localization in UWSNs. The technique utilizes Intrusion Detection System (IDS) methods to calculate the trust values of secondary-anchor nodes (SNs), ensuring only the most trusted SNs are selected. Furthermore, a digital signature algorithm secures the Received Signal Strength (RSS) data from tampering. The method incorporates a hybrid approach combining Angle of Arrival (AoA) and RSS ranging techniques to achieve robust localization. Simulation results demonstrate that JTDSA improves localization accuracy by 4% compared to exist ing methods, achieving an overall accuracy of 99.2%. The technique also reduces packet drop by 35% and enhances attack detection accuracy by 3%. These results indicate that the proposed JTDSA technique offers an effective solution for secure and reliable localization in UWSNs
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
    Monitoring ambient parameters in the IoT precision agriculture scenario: An approachto sensor selection and hydroponic saffron cultivation
    (Sensors, 2022) Kour, Kanwalpreet; Gupta, Deepali; Gupta, Kamali; Anand, Divya; Elkamchouchi, Dalia H.; Goyal, Nitin; Ibrahim, Muhammad
    The world population is on the rise, which demands higher food production. The reduction in the amount of land under cultivation due to urbanization makes this more challenging. The solution to this problem lies in the artificial cultivation of crops. IoT and sensors play an important role in optimizing the artificial cultivation of crops. The selection of sensors is important in order to ensure a better quality and yield in an automated artificial environment. There are many challenges involved in selecting sensors due to the highly competitive market. This paper provides a novel approach to sensor selection for saffron cultivation in an IoT-based environment. The crop used in this study is saffron due to the reason that much less research has been conducted on its hydroponic cultivation using sensors and its huge economic impact. A detailed hardware-based framework, the growth cycle of the crop, along with all the sensors, and the block layout used for saffron cultivation in a hydroponic medium are provided. The important parameters for a hydroponic medium, such as the concentration of nutrients and flow rate required, are discussed in detail. This paper is the first of its kind to explain the sensor configurations, performance metrics, and sensor-based saffron cultivation model. The paper discusses different metrics related to the selection, use and role of sensors in different IoT-based saffron cultivation practices. A smart hydroponic setup for saffron cultivation is proposed. The results of the model are evaluated using the AquaCrop simulator. The simulator is used to evaluate the value of performance metrics such as the yield, harvest index, water productivity, and biomass. The values obtained provide better results as compared to natural cultivation.
  • No Thumbnail Available
    Item
    Privacy-Preserving and Collaborative Federated Learning Model for the Detection of Ocular Diseases
    (2025) Goyal, Nitin
    Abstract Ocular diseases significantly impact the health of the public globally. According to the World Health Organization (WHO) reports, at least 1 billion people suffer from near or distance vision impairment that could have been prevented or has yet to be addressed. These conditions cause difficulty in living a healthy lifestyle and impair individual quality of life. The article explores the application of federated learning in detecting two vision-threatening ocular diseases- diabetic retinopathy and diabetic macular edema. A federated learning framework enhances the technological capabilities of artificial intelligence that leverages decentralised data sources without creating data banks to maintain privacy. The methodology implements a federated learning environment with 2, 3, and 4 clients, using MobileNetV2 as the backbone deep learning model. The model is trained on a composite of 2 datasets procured from the Kaggle repository, comprising coloured fundus images labelled for diabetic retinopathy, diabetic macular edema, and normal cases. The federated learning process involves training at the client end to build client models called local models. The clients in a federated learning system only share updates regarding their local models. The original data is never shared with a central server. The server integrates these local models into the central global models using aggregation strategies such as FedAvg, FedProx, etc. Performance metrics, including prediction accuracy, class-wise accuracy, precision, recall, and F1 score, are calculated across 30 communication rounds. The results demonstrate that the federated learning model achieves an average prediction accuracy of 96%, and a class-wise accuracy of 100% in detecting diabetic macular edema and diabetic retinopathy. The high performance of the federated learning system highlights the significance of federated learning as a viable solution for ocular disease detection while ensuring data privacy.
  • Thumbnail Image
    Item
    System for Refrigeration
    (2023-01-24) Goyal, Nitin
  • Thumbnail Image
    Item
    A transfer learning-based artificial intelligence model for leaf disease assessment
    (Sustainability, 2022) Gautam, Vinay; Trivedi, Naresh K.; Singh, Aman; Mohamed, Heba G.; Goyal, Nitin; Kaur, Preet
    The paddy crop is the most essential and consumable agricultural produce. Leaf disease impacts the quality and productivity of paddy crops. Therefore, tackling this issue as early as possible is mandatory to reduce its impact. Consequently, in recent years, deep learning methods have been essential in identifying and classifying leaf disease. Deep learning is used to observe patterns in disease in crop leaves. For instance, organizing a crop’s leaf according to its shape, size, and color is significant. To facilitate farmers, this study proposed a Convolutional Neural Networksbased Deep Learning (CNN-based DL) architecture, including transfer learning (TL) for agricultural research. In this study, different TL architectures, viz. InceptionV3, VGG16, ResNet, SqueezeNet, and VGG19, were considered to carry out disease detection in paddy plants. The approach started with preprocessing the leaf image; afterward, semantic segmentation was used to extract a region of interest. Consequently, TL architectures were tuned with segmented images. Finally, the extra, fully connected layers of the Deep Neural Network (DNN) are used to classify and identify leaf disease. The proposed model was concerned with the biotic diseases of paddy leaves due to fungi and bacteria. The proposed model showed an accuracy rate of 96.4%, better than state-of-the-art models with different variants of TL architectures. After analysis of the outcomes, the study concluded that the anticipated model outperforms other existing models.