Bhupesh Dewangan

Dr. Bhupesh Dewangan

University of Petroleum and Energy Studies, Dehradun
orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
9 February 2023 in Electronics
Verified
Journal Article
Open access
PDF file
MAMonika Arya
Monika Arya
HSHanumat Sastry
Hanumat Sastry
BDBhupesh Dewangan
Bhupesh Dewangan
MRMohammad Khalid Imam Rahmani
Mohammad Khalid Imam Rahmani
SBSurbhi Bhatia
Surbhi Bhatia
MBMariyam Aysha Bivi
Mariyam Aysha Bivi
Vehicular networks improve quality of life, security, and safety, making them crucial to smart city development. With the rapid advancement of intelligent vehicles, the confidentiality and security concerns surrounding vehicular ad hoc networks (VANETs) have garnered considerable attention. VANETs are intrinsically more vulnerable to attacks than wired networks due to high mobility, common network medium, and lack of centrally managed security services. Intrusion detection (ID) servers are the first protection layer against cyberattacks in this digital age. The most frequently used mechanism in a VANET is intrusion detection systems (IDSs), which rely on vehicle collaboration to identify attackers. Regrettably, existing cooperative IDSs get corrupted and cause the IDSs to operate abnormally. This article presents an approach to intrusion detection based on the distributed federated learning (FL) of heterogeneous neural networks for smart cities. It saves time and resources by using the most efficient intruder detection approach. First, vehicles use a federated learning technique to develop local, deep learning-based IDS classifiers for VANET data streams. They then share their locally learned classifiers upon request, significantly reducing communication overhead with neighboring vehicles. Then, an ensemble of federated heterogeneous neural networks is constructed for each vehicle, including locally and remotely trained classifiers. Finally, the global ensemble model is again shared with local devices for their updating. The effectiveness of the suggested method for intrusion detection in VANETs is evaluated using performance indicators such as attack detection rates, classification accuracy, precision, recall, and F1 scores over a ToN-IoT data stream. The ID model shows 0.994 training and 0.981 testing accuracy.

21 September 2021 in GeoJournal
Verified
Journal Article
Closed access
PDF file
DJDev Joshi
Dev Joshi
APAtul Kumar Patidar
Atul Kumar Patidar
AMAbhipshit Mishra
Abhipshit Mishra
AMAditya Mishra
Aditya Mishra
SASomya Agarwal
Somya Agarwal
Tanupriya ChoudhuryTanupriya Choudhury
Tanupriya Choudhury

21 February 2020 in Software: Practice and Experience
Verified
Journal Article
Closed access
PDF file
BDBhupesh Dewangan
Bhupesh Dewangan
AAAmit Agarwal
Amit Agarwal
TCTanupriya Choudhury
Tanupriya Choudhury
APAshutosh Pasricha
Ashutosh Pasricha
SSSuresh Chandra Satapathy
Suresh Chandra Satapathy

10 October 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
PSPreeti Sirohi
Preeti Sirohi
FAFahd N. Al-Wesabi
Fahd N. Al-Wesabi
HAHaya Mesfer Alshahrani
Haya Mesfer Alshahrani
PMPiyush Maheshwari
Piyush Maheshwari
AAAmit Agarwal
Amit Agarwal
Tanupriya ChoudhuryTanupriya Choudhury
Tanupriya Choudhury
The growing demand for cloud technology brings several cloud service providers and their diverse list of services in the market, putting a challenge for the user to select the best service from the inventory of available services. Therefore, a system that understands the user requirements and finds a suitable service according to user-customized requirements is a challenge. In this paper, we propose a new cloud service selection and recommendation system (CS-SR) for finding the optimal service by considering the user’s customized requirements. In addition, the service selection and recommendation system will consider both quantitative and qualitative quality of service (QoS) attributes in service selection. The comparison is made between proposed CS-SR with three existing approaches analytical hierarchy process (A.H.P.), efficient non-dominated sorting-sequential search (ENS-SS), and best-worst method (B.W.M.) shows that CR-SR outperforms the above approaches in two ways (i) reduce the total execution time and (ii) energy consumption to find the best service for the user. The proposed cloud service selection mechanism facilitates reduced energy consumption at cloud servers, thereby reducing the overall heat emission from a cloud data center.

1 January 2021 in International Journal of Grid and High Performance Computing
Verified
Journal Article
Closed access
PDF file
BDBhupesh Dewangan
Bhupesh Dewangan
VMVenkatadri M.
Venkatadri M.
AAAmit Agarwal
Amit Agarwal
APAshutosh Pasricha
Ashutosh Pasricha
Tanupriya ChoudhuryTanupriya Choudhury
Tanupriya Choudhury
In cloud computing, applications, administrations, and assets have a place with various associations with various goals. Elements in the cloud are self-sufficient and self-adjusting. In such a collaborative environment, the scheduling decision on available resources is a challenge given the decentralized nature of the environment. Fault tolerance is an utmost challenge in the task scheduling of available resources. In this paper, self-healing fault tolerance techniques have been introducing to detect the faulty resources and measured the best resource value through CPU, RAM, and bandwidth utilization of each resource. Through the self-healing method, less than threshold values have been considering as a faulty resource and separate from the resource pool. The workloads submitted by the user have been assigned to the available best resource. The proposed method has been simulated in cloudsim and compared the multi-objective performance metrics with existing methods, and it is observed that the proposed method performs utmost. Request access from your librarian to read this article's full text.

30 September 2020 in Revue d'intelligence artificielle
Verified
Journal Article
Open access
BDBhupesh Dewangan
Bhupesh Dewangan
AJAnurag Jain
Anurag Jain
Tanupriya ChoudhuryTanupriya Choudhury
Tanupriya Choudhury
Resource optimization is cost effective process in cloud. The efficiency of load balancing completely depends on how the infrastructure is utilizing. As per the current study, the resource optimization techniques are very costly and taking more convergence time to execute the task and load distribution among different virtual machines (VM). The objective of this paper is to develop a hybrid optimization algorithm to find the best virtual machine based on their fitness values and schedule different task to the fittest VM so that each task should get complete on time, and system can utilize the VM as well. The proposed algorithm is hybrid version of genetic (GA), ant-colony (Aco), and particle-swarm (Pso) algorithms, which is implemented and tested in amazon web service and compared with existing algorithms based on VM utilization, completion time, and cost. The proposed hybrid system genetic-aco-pso based algorithm (GAP) perform utmost while comparing with the existing systems.