VIRAL Vinod KAPADIA

Dr. VIRAL Vinod KAPADIA

orcid

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 June 2020 in Cybernetics and Information Technologies
Verified
Journal Article
Open access
PDF file
VBVirendra Barot
Virendra Barot
VKVIRAL Vinod KAPADIA
VIRAL Vinod KAPADIA
ANAnd
And
Sharnil PandyaSharnil Pandya
Sharnil Pandya
Air pollution has emerged as a major concern of the current century. In recent times, fellow researchers have conducted numerous researches in the area of air quality monitoring. Still, air quality monitoring remains an open research area due to various challenges such as sophisticated topology design, privacy and security, power backup, large memory requirements and deployment of such systems at resource-constrained sites. The proposed research work is an attempt to address the issues of communication topology design, assessment of the Quality of Service (QoS) levels against accuracy, sensing throughput and power consumption optimization. In the undertaken work, the proposed IoT based Air Quality Monitoring system has been deployed at indoor and outdoor sites to measure air quality parameters such as PM10, PM2.5, carbon monoxide, temperature and humidity. The proposed system is also tested at variety of quality of service levels at the indoor and outdoor sites. The conducted experiments have also recorded accuracy in terms of reliable delivery of the messages under employed protocol.

1 March 2022 in Cybernetics and Information Technologies
Verified
Journal Article
Open access
PDF file
VBVirendra Barot
Virendra Barot
VKVIRAL Vinod KAPADIA
VIRAL Vinod KAPADIA
Air pollution has increased worries regarding health and ecosystems. Precise prediction of air quality parameters can assist in the effective action of air pollution control and prevention. In this work, a deep learning framework is proposed to predict parameters such as fine particulate matter and carbon monoxide. Long Short Term Memory (LSTM) neural network-based model that processes sequences in forward and backward direction to consider the influence of timesteps in both directions is employed. For further learning, unidirectional layers’ stacking is implemented. The performance of the model is optimized by fine-tuning hyperparameters, regularization techniques for overfitting resolution, and various merging options for the bidirectional input layer. The proposed model achieves good optimization and performs better than the simple LSTM and a Recurrent Neural Network (RNN) based model. Moreover, an attention-based mechanism is adopted to focus on more significant timesteps for prediction. The self-attention approach improves performance further and works well especially for longer sequences and extended time horizons. Experiments are conducted using real-world data collected, and results are evaluated using the mean square error loss function.

20 May 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
AAAli Saeed Almuflih
Ali Saeed Almuflih
DVDhairya Jagdipchandra Vyas
Dhairya Jagdipchandra Vyas
VKVIRAL Vinod KAPADIA
VIRAL Vinod KAPADIA
MOHAMED RAFIK NOOR MOHAMED QURESHIMOHAMED RAFIK NOOR MOHAMED QURESHI
MOHAMED RAFIK NOOR MOHAMED QURESHI
KQKarishma M. Qureshi
Karishma M. Qureshi
EMElaf Abdullah Makkawi
Elaf Abdullah Makkawi
In machine learning (ML), adversarial attack (targeted or untargeted) in the presence of noise disturbs the model prediction. This research suggests that adversarial perturbations on pictures lead to noise in the features constructed by any networks. As a result, adversarial assaults against image categorization systems may present obstacles and possibilities for studying convolutional neural networks (CNNs). According to this research, adversarial perturbations on pictures cause noise in the features created by neural networks. Motivated by adversarial perturbation on image pixel attacks observation, we developed a novel exploit feature map that describes adversarial attacks by performing individual object feature-map visual description. Specifically, a novel detection algorithm calculates each object’s class activation map weight and makes a combined activation map. When checked with different networks like VGGNet19 and ResNet50, in both white-box and black-box attack situations, the unique exploit feature-map significantly improves the state-of-the-art in adversarial resilience. Further, it will clearly exploit attacks on ImageNet under various algorithms like Fast Gradient Sign Method (FGSM), DeepFool, Projected Gradient Descent (PGD), and Backward Pass Differentiable Approximation (BPDA).

8 October 2021
Verified
Book Chapter
Closed access
PDF file
KPKhushi Patel
Khushi Patel
VKVIRAL Vinod KAPADIA
VIRAL Vinod KAPADIA

27 July 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
AAAli Saeed Almuflih
Ali Saeed Almuflih
KPKhushi Patel
Khushi Patel
VKVIRAL Vinod KAPADIA
VIRAL Vinod KAPADIA
MOHAMED RAFIK NOOR MOHAMED QURESHIMOHAMED RAFIK NOOR MOHAMED QURESHI
MOHAMED RAFIK NOOR MOHAMED QURESHI
NANaif Almakayeel
Naif Almakayeel
RMRabia Emhamed Al mamlook
Rabia Emhamed Al mamlook
Across the globe, wireless devices with Internet facilities such as smartphones and tablets have become essential assets for communication and entertainment alike for everyday life for millions of people, which increases the network traffic and the demand for low-latency communication networks. The fourth-generation (4G)/long-term evolution (LTE)/ fifth-generation (5G) communication technology offers higher bandwidth and low latency services, but resource utilization and resiliency cannot be achieved, as transmission control protocol (TCP) is the most common choice for most of the state-of-art applications for the transport layer. An extension of TCP—multipath TCP (MPTCP)—offers higher bandwidth, resiliency, and stable connectivity by offering bandwidth aggregation and smooth handover among multiple paths. However, MPTCP uses multiple disjointed paths for communication to offer multiple benefits. A breach in the security of one of the paths may have a negative effect on the overall performance, fault-tolerance, robustness, and quality of service (QoS). In this paper, the research focuses on how MPTCP options such as MP_CAPABLE, ADD_ADDR, etc., can be used to exploit the vulnerabilities to launch various attacks such as session hijacking, traffic diversion, etc., to compromise the availability, confidentiality, and integrity of the data and network. The probable security solutions for securing MPTCP connections are analyzed, and the secure key exchange model for MPTCP (SKEXMTCP) based on identity-based encryption (IBE) is proposed and implemented. The parameters exchanged during the initial handshake are encrypted using IBE to prevent off-path attacks by removing the requirement for key exchange before communication establishment by allowing the use of arbitrary strings as a public key for encryption. The experiments were performed with IBE and an elliptic curve cryptosystem (ECC), which show that IBE performs better, as it does not need to generate keys while applying encryption. The experimental evaluation of SKEXMTCP in terms of security and performance is carried out and compared with existing solutions.

1 March 2009
Verified
Conference Paper
Closed access
VKVIRAL Vinod KAPADIA
VIRAL Vinod KAPADIA
DTDarshak G Thakore
Darshak G Thakore
In Today's era distributed system is gaining too much popularity and at the same time there are many issues related to distributed system like memory management, message passing, consistency, replicas, speed of operations, transparency and many more. We would like to focus on garbage collection in distributed system which is also a very important issue. Here we are considering the train algorithm given by Hudson & Moss and focus that if we use distributed garbage collector's train algorithm in server-client application, the performance can be improved..