Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
8 results
Search Results
Item Early Detection and Classification of Zero-Day Attacks in Network Traffic Using Convolutional Neural Network(Springer Science and Business Media Deutschland GmbH, 2024) Singh, M.P.; Singh, V.P.; Gupta, M.In a Zero-Day cyberattack, attackers exploit a software vulnerability for which the software vendor is unaware or has not released a patch. This can make it difficult for organizations to protect their systems until a patch or mitigation is developed. To stay ahead of these evolving cyber threats, it’s critical to keep up to date with the latest threat information and to remain vigilant. Traditional methods for detecting and classifying zero-day attacks often require session-wide features, which can be challenging to implement. This paper presents a novel approach for detecting and classifying Zero-Day attacks in network traffic. Specifically, we present a framework composed of a 1D Convolutional Neural Network (1D-CNN), which involves minimal preprocessing and directly leverages raw network data as byte sequences to learn features, eliminating the need for complex feature extraction. To test the effectiveness of our proposed approach, publicly available network traffic datasets encompassing various malware families are used. Results show that the proposed approach is significantly effective in detecting and classifying Zero-Day attacks, empowering organizations to combat evolving cyber threats. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Item Improvement in Performance of InAs Surface Quantum Dot Heterostructure-Based H2S Gas Sensor by Introducing Buried Quantum Dot Layer(Institute of Electrical and Electronics Engineers Inc., 2023) Mantri, M.R.; Panda, D.P.; Punetha, D.; Pandey, S.K.; Singh, V.P.; Pandey, S.K.; Chakrabarti, S.In this work, we have demonstrated InAs surface quantum dot (SQD)-based H2S gas sensors. The epitaxial growth of the strain-coupled and uncoupled InAs/GaAs QD heterostructures is done using the solid-source molecular beam epitaxy (MBE) tool. For both types of heterostructures, the coverage of the InAs monolayer (ML) for the SQD layer varies from 0.9 to 2 ML. The ML coverage of the buried quantum dots (BQDs) layer for the coupled heterostructures is kept constant (2.7 ML). The atomic force microscopy (AFM) results demonstrated that the coupled heterostructures have higher quantum dot (QD) density in the SQDs layer in comparison to the uncoupled one due to strain propagation from the BQDs toward the SQD layer. The sensor fabricated using the coupled heterostructure with 2 ML SQDs has demonstrated better performance than the uncoupled one for various concentrations (1-1000 ppm) of hydrogen sulfide (H 2S) gas due to inter-dot carrier tunneling between BQDs and SQDs layer. The coupled InAs gas sensor showed the best sensing properties at room temperature (45.9% sensor response at 100 ppm H2S ). We have demonstrated the selectivity of the sensor toward H 2S among various target gases like CO, CO2 , N2O , and NO 2 and the stability over a longer period of time with only 3% deviation (within acceptable limit). These findings have the potential to promote the fabrication of high-performance gas sensors using SQDs-based coupled heterostructures. © 2001-2012 IEEE.Item Security in 5G Network Slices: Concerns and Opportunities(Institute of Electrical and Electronics Engineers Inc., 2024) Singh, V.P.; Singh, M.P.; Hegde, S.; Gupta, M.Network slicing has emerged as a cornerstone technology within the 5G ecosystem, enabling efficient resource allocation, service customization, and support for various applications. Its ability to deliver Network-as-a-Service (NaaS) brings a new paradigm of adaptable and efficient network provisioning. However, with the diversification of services and the increasing complexity of network infrastructures, a simultaneous rise in security vulnerabilities becomes evident. These flaws go beyond the limitations of conventional network security and affect various aspects of network slice (NS) implementation and management. The limitations of traditional security, such as static policies, single point of failure, and challenges in effectively securing network slicing deployments, underscore the need to explore security measures tailored to the dynamic nature of 5G networks. To ensure the robust security of 5G networks, it is essential to consider various security concerns such as isolation, authentication, and authorization. Furthermore, dynamic orchestration and inter-slice communication security challenges must be proactively tackled. The security concerns related to 5G networks must be addressed comprehensively to ensure the safe and secure operation of the network. Our survey paper goes into these complex security issues, providing an in-depth and systematic review of the various contexts in which they emerge. We have identified the five most vulnerable areas in Network Slicing: Slice-Lifecycle, Communication type slice uses, Technologies used to provide service, Management threats, and End Devices utilized in service. Apart from threats in these vulnerable areas, we also discussed a few generous attacks that can be launched to disrupt network-slicing services. Furthermore, this study is a valuable resource for evaluating the current state of research efforts in this domain, contributing to the ongoing enhancement of security measures and the overall robustness of network-slicing technology. In doing so, we aim to ensure the secure and sustainable evolution of 5G networks as they become increasingly integral to our digital infrastructure. © 2013 IEEE.Item Graphene-Oxide-Coated Flexible Fabric Antenna Sensor for Contact-Free H2O Sensing(Institute of Electrical and Electronics Engineers Inc., 2024) Singh, V.P.; Kandasamy, K.; Rahman, M.R.This article presents a flexible antenna sensor on denim jeans fabric for H2O sensing. A small defected ground structure is developed on the copper ground plane of the antenna. Graphene oxide (GO) is coated on the defected ground for sensor applications. The flexible antenna sensor with GO coated on a defected ground plane resonates at 5.52 GHz. The resonance frequency of the proposed antenna changes linearly with the H2O content. The frequency sensitivity of the denim fabric antenna sensor is ≈203 MHz/10% relative humidity (RH). The denim fabric antenna sensor has a measured phase sensitivity of 40° per 10% RH. The flexible antenna sensor gives a linearity of 98.28%. The designed antenna sensor can distinguish between fresh and dried grapes by H2O detection. © 2001-2012 IEEE.Item A Flexible and Biodegradable Graphene Oxide Antenna Sensor for Monitoring Subsoil Health(American Chemical Society, 2024) Singh, V.P.; Kandasamy, K.; Rahman, M.R.In this paper, a flexible graphene oxide-based antenna sensor is designed on the biodegradable substrate. It resonates at a frequency of 3.92 GHz. The flexible sensor is used in assessing water content in sandy loam subsoil near the Arabian Sea from the Western Ghats in India. The volumetric water content (VWC) of dry soil varies between approximately 0.06 VWC m3 m-3 to 0.7 VWC m3 m-3. The antenna sensor exhibits a linearity of 92.02% and a sensitivity of 402.6 MHz/VWC(m3 m-3). The limit of detection (LOD) and limit of quantification (LOQ) of the antenna sensor are 0.75 VWC and 2.5 VWC m3 m-3. The sensor is beneficial for examining soil water to enhance crop productivity for precision agriculture. © 2024 American Chemical Society.Item Fortifying 5G network slices using a trust-based secure federated learning framework for attack detection and classification(Springer Science and Business Media Deutschland GmbH, 2025) Singh, V.P.; Singh, M.P.; Hegde, S.The rapid evolution of 5G has revolutionized communication by offering high-speed connectivity and supporting various applications. An essential feature of 5G networks is network slices, which enable the creation of multiple virtualized and independent networks on a shared physical infrastructure to provide a dynamic range of services for specific use cases. However, this flexibility also poses significant security challenges, such as Denial of Service (DoS), Distributed Denial of Service (DDoS), Injection, Malware, and Man-in-the-middle (MITM), to network slices. In addition, existing Machine Learning (ML) and Deep Learning (DL) based approaches cannot adapt to network slices’ distributed and dynamic nature, posing privacy threats. Unlike conventional methods, Federated Learning (FL) presents a more advanced alternative with enhanced security and privacy. However, FL aggregation processes remain vulnerable to several attacks, including model poisoning, data poisoning, and Byzantine attacks. Addressing them is essential for unlocking FL’s complete potential. This paper proposes a trust-based client selection technique to secure FL by ensuring that only trusted, non-malicious clients contribute to global model development. In addition, our proposed secure FL framework uses the ResNet-18 Convolutional Neural Network (CNN) to detect and classify attacks in network slices, achieving 97.36% accuracy in non-malicious environments. The proposed approach significantly outperforms in the presence of 60% and 70% malicious clients, and demonstrates 93.35% and 56.38% accuracy, respectively. These results highlight the effectiveness of our secure FL framework for detecting and classifying attacks in network slices, even in the presence of malicious clients. Furthermore, an experimental analysis on the Edge-IIoT dataset demonstrates the generalizability and robustness of the proposed framework. © Springer-Verlag GmbH Germany, part of Springer Nature 2025.Item Graphene Oxide Assisted Humidity Sensing Antenna Sensor(Korean Institute of Electrical and Electronic Material Engineers, 2025) Singh, V.P.; Kandasamy, K.; Rahman, M.R.Abstract: In this paper, a graphene oxide (GO) coated substrate material is electrically characterized. The permittivity, loss tangent, permeability, and resistance are measured. A planar antenna with slots is designed on an RT/Duroid 5880 substrate. A thin layer of graphene oxide is coated on the identical slots of the planar antenna. The GO-assisted antenna sensor resonates at 2.51 GHz frequency. The graphene oxide-based antenna sensor performs contact-free sensing of relative humidity. The frequency sensitivity is approximately 17.5 MHz/10%RH. The design sensor gives a 93.50% linearity. The limit of detection is 30.55 RH%. Graphic Abstract: (Figure presented.) © The Korean Institute of Electrical and Electronic Material Engineers 2024.Item Enhancing the Security of SDN in 5G: A Hybrid Feature Selection Based Ensemble Machine Learning Framework for Classification of Cyber-Attacks(Springer, 2025) Singh, M.P.; Haimashreelakshmi; Singh, V.P.; Gupta, M.In recent years, the adoption of 5G has significantly increased due to its numerous benefits, including high availability, lower latency, improved reliability, and high performance. To manage packet flow, 5G relies on Software-Defined Networking (SDN) that employs software controllers and Application Programming Interfaces (APIs) to route packets and communicate with the hardware, providing advantages like high efficiency, low cost, and dependability. However, due to centralized control, SDN controllers are vulnerable to various cyber-attacks, including Distributed Denial of Service (DDoS), Denial of Service (DoS), Password Brute Forcing, Web Attacks, etc. This paper proposes a framework that comprises a hybrid feature selection method and an ensemble machine learning model. The proposed ensemble model combines the strengths of three different machine learning (ML) classifiers to create a voting classifier for classifying traffic in SDN. Additionally, the optimal value for the hyperparameters of each classifier is obtained through hyperparameter tuning. Finally, the experimental analysis of the proposed model using the InSDN dataset shows 99.96% accuracy, highlighting the proposed model’s effectiveness in addressing the limitations of the existing approaches and detecting multiple attacks in the SDN context. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.