Dr. Suraj Malik

Scopus Publications

Enhanced radial base deep learning algorithm for predicting multimedia security issues
V. Haripriya, Mohan Vishal Gupta, Nikita Nadkarni, Suraj Malik, Aditya Yadav, and Apoorva Joshi
IOS Press
From online social networks to life-or-death security systems, multimedia files (photos, movies, and audio recordings) have grown common in today’s digital culture. Protecting people, businesses and infrastructure requires strict adherence to the encryption and decryption of multimedia data. We suggested an Ensemble Whale Optimized Recurrent Neural Network (EWO-RNN)used in this study to overcome these issues. With the help of this study, multimedia security will be evaluated in more accurate and comprehensive manner. Smarter decisions and proactive security measures may follow as a result of this. To increase the system quality and the overall performance, the collected data is pre-processed for normalized data by using Min-Max Normalization. Pre-processed data is extracted by using Kernel Principle Component Analysis (K-PCA). The EWO-RNN evaluates the effectiveness and efficiency of an approach by analyzing the performance of Accuracy (97.85%), Precision (92.2%), F1-score (96.1%), Mean Square Error (MSE) (0.086), Root Mean Square (RMSE) (0.12%) and Sensitivity (95%). The Enhanced Radial Base Deep Learning Algorithm for Predicting Multimedia Security Issues proposes a solution with improved resilience, accuracy, generalization, and decision-making capabilities. In a dynamic and evolving digital environment, this makes the algorithm a viable tool for multimedia security assessments.

Development Arena through Cloud Computing Technology
Archana Jain, Nidhi Bansal, Mahima Shankar Pandey, Gaurav Kumar, Manoj Kumar Sharma, Suraj Malik, and Santosh Kumar Maurya
IEEE
Latest measurements correlated to the cloud computing technology, found to be very unreliable. For smooth conduction of cloud technology, the report is getting more than 100 values i.e., being added to the cost of the essential component that has been prepared for succession of the technology. Moreover, with the care of cloud computing technology, very intelligent operational programing has been into running, yet the real parameter of performance measurement is showing the actual output. With tremendous literature survey and feedback, the paper has been able to lay out logical terms about the services availed from cloud side. Trendy services involve trust and reliable factors in terms of benefits accruing from the customer side to the service provider. By offering the combined efforts of public and private cloud, business drivers are moving forward with the freedom to deliver their most effective service to the users and are also able to achieve higher spend.

Massive MIMO based beamforming by optical multi-hop communication with energy efficiency for smart grid IoT 5G application
Asha Rajiv, Pankaj Kuamr Goswami, Rajesh Gupta, Suraj Malik, Usha Chauhan, and Anil Agarwal
Springer Science and Business Media LLC

Optical system based data classification for diabetes retinopathy detection using machine language with artificial intelligence
Suraj Malik, S. Srinivasan, Chandra Shekhar Rajora, Sachin Gupta, Mohammed Mujeer Ulla, and Neeraj Kaushik
Springer Science and Business Media LLC

Virtual machine tree task scheduling for load balancing in cloud computing
Santosh Kummar Maurya, Suraj Malik, and Neeraj Kumar
Institute of Advanced Engineering and Science
The increasing number of publications towards cloud computing proves that much research and development has been done, especially for task scheduling. Organizations are eager to get more customized technology to run the most smoothly in the provision of visual cloud services for fruity users. As the circumstances of Covid indicate to technology that everyone should run digitally, the workload on machines increased. For workload solutions, organizations are trying to balance the situation with the successful operation of cloud services to use appropriate services/resources. Nevertheless, the issues are still to be resolved by researchers, so we respect all my friends who are putting a lot of effort into developing new techniques. A proposed paper is showing a new collation with the load balancing factor by implementing quality of service (QoS) and virtual machine tree (VMT). A CloudSim toolkit will then be used to compare them. A tree structure graph is included in the VMT algorithm to schedule tasks with the appropriate distribution on each machine. The QoS algorithm performs the task of scheduling based on the service required by the user with the best quality and satisfies the user.

The Importance of Data Analysis in the Field of Ultra Smart Grid Using an Ultra Smart Grid
Humera Aqeel, Suraj Malik, Santosh Kumar, G. Ravivarman, G. Premananthan, and Suruchi Gaurav Dedgaonkar
IEEE
Automated manufacturing networks now place a greater emphasis on data processing. With the widespread implementation of smart feet and devices, a digital layer has now been added to the traditional energy manufacturing and delivery network for data gathering, storing, and analyzation. This growth is caused by the growth of communication and information new tech. This article provides an introduction to big data analytics and related national grid uses. Before demonstrating the reason and possible benefits of adopting sophisticated analytics in power buildings, the characteristics of big data, microgrids, and massive amounts of data gathering are first addressed. There is also discussion of the fundamental ideas and methods used in normal business intelligence for common issues. The primary focus of this article is on sophisticated uses of various data analyses in smart networks. In the age of big data, the current energy strategy can gain many advantages by utilizing vast amounts of data out from power source, aculeate system, location computer system, etc. This will also improve consumer experience but instead social equity.

The Technical Importance of Accurate Diagnosis Using AI and ML Algorithms in the Field of Radioloy: A Medical Application
Shashi Mehrotra, Abishek Agarhari, Suraj Malik, K. Yuvaraj, P. Jeevananthan, and Pradnya Mehta
IEEE
Due to the accessibility of sizable databases, significant advancements in processing power, and novel approaches to deep learning, excitement in applications of AI-such as those involving imaging-is high and increasing quickly on a global scale. There are numerous possibilities and hurdles for the visualization group aside from the invention of fresh Methodologies per se, such as the creation of a widespread designations, improved methods for sharing image data, and guidelines for vindicating application use across different imaging social media channels and patient groups. By reporting malicious or favorable cases for honest review, intelligence services may aid pharmacists in prioritizing their to-do lists. They are additionally capable of helping extract brain mri data from photos that cannot be seen visually, done more easily the diagnostic as well as prognostlc value of data sets. There have been prophecies that doctors will become obsolete due to AI. This problem has been exaggerated, and doctors are considerably more likely to effectively integrate AI techniques into about their operations. The current shortages of technological know-how and even processing capacity will be handled in time, and internet connectivity options too can help. AI in radiography will be successful if it adds value through improvements in patient results, quicker response times, greater diagnosis confidence, and doctors' satisfaction with their lives at employment. A fresh and innovative collection of approaches for studying picture data are provided by AI. These novel avenues will be investigated by specialists, who will probably take the lead in AI's uses to medicine.

A New Method of Image Steganography Technique Based on Fingerprint with Qr-Code Using Watermarking Technique
Manoj Kumar Sharma, Nidhi Bansal, Suraj Malik, Gaurav Kumar, and Archana Jain
IEEE
This research paper based on new steganography technology with the using of fingerprint and QR code using watermarking technique. Now's a day digital technology and image steganography more useful for secure data. User and developer both no perfect secure data. According to digital technology biometric features is good for secure data. Biometric features are the part of human body which has excellent structure to be determined perfectly. In this paper it is proposed to use biometric fingerprint with QR-code by image steganography and cryptography using watermarking technique and produce more secure results comparison to other techniques.

Application of Internet of Things and Machine Learning in Improving Supply Chain Financial Risk Management System
Kulbir Singh, Richa, Prakash Divakaran, Vandana Mishra Chaturvedi, Pradeep Kumar, and Suraj Malik
IEEE
This study looks into the significant operational risk, information asymmetry, and modal lag problems that are currently present in the market for supply chain financing between banks and enterprises. Policy suggestions will be made based on the investigation’s results. This study set out to define supply chain finance and illustrate its core goal. The first and most crucial step was this. From a theoretical perspective, this study looked into the conceptual foundations of financial risk management in supply networks. In an effort to ascertain whether or not they could assist in the process of avoiding potential risks during the actual production of the product, the self-compensated trade finance theory, the information asymmetry theory, and the use of principle agents were all put to the test. After that, an investigation of the deficiencies of the operational strategies and associated business practices of the three primary supply chain finance models was carried out. On the basis of this, operational risks were found, and supply chain operations’ financial elements as well as risk sources were investigated. Creating a strategy for risk mitigation is the first step for a business. The next phase is to implement this plan as soon as it is practical, and you should continuously assess its effectiveness. In the context of managing pandemics and natural disasters, a significant amount of input from the government should be incorporated into the process of giving aid in the form of both financial and nonfinancial resources. This is the case since actively participating in one’s local community is one of the best methods to create a strong society.

Advancing Wireless Communication: A Comprehensive Study on Multi-User Mimo-Ofdm Techniques
Praveen Singh, Rahul Vishnoi, Suraj Malik, M. Vigenesh, S. Arul Antran Vijay, and Monali Ravindra Borade
IEEE
A actually massive settings is created when horizontal frequency division packet switching, a common technique for high-rate wireless data transport, is integrated with transceivers at the transmitting and receiving ends to increase spectral efficiency and/or to boost system power on time-variant and intensity touchpoints. The multiple access future research directions in design process are investigated in this article, which would include measured data and simulation, input signal spectrum sensing methods employing intuitive arrays, space-time methods, error - correcting computer programming methods, protocol and datagram style, or signal computation for frequency and time standardization, spectrum sensing, and channel locating in frameworks. The article also takes into account an open-source hardware solution.

IoT Applications in Blockchain Technology
Ajay Sudhir Bale, Ankur Biswas, Suraj Malik, Enjula Uchoi, Sabhyata Uppal Soni, and Arun Soni
IEEE
Securing the data through blockchain might be crucial to enhancing the IoT network, since data is an especially important part of the system. Secure storage of the information generated by the IoT network's intelligent gadgets is made possible via the integration of blockchain technology. For blockchain to be widely used in the IoTs, it must first conquer technological and financial barriers. To make the most of the IoT, blockchain technology may require some tweaking. The upcoming Internet of Things (IoT) technical breakthrough could involve this. This work emphasizes on bringing on board the recent trends of IoT applications utilizing Blockchain technology. The challenges faced in doing so are also outlined. We're hoping this work might help elucidate the state of blockchain studies and provide a road map for what comes next.

Recent Scientific Achievements and Developments in Software Defined Networking: A Survey
Ajay Sudhir Bale, Hamsalekha R, Bishwajeet Prakash, Pidikiti Surendra Babu, Rashmi Gupta, and Suraj Malik
IEEE
Application devices or application programming interfaces (APIs) that are connected with the channel's core embedded system and regulate congestion is known as software-defined networking (SDN). Architecture is the primary distinction among SDN and conventional connectivity. Traditional networking is hardware-based, whereas SDN is software-based. Compared to conventional networks, SDN is far more versatile since the routing protocol is software-based. Without adding extra equipment, it enables managers to handle the connection, alter customization options, supply resources, and boost internet bandwidth from a single user interface. There are a lot of developments in SDN which have enabled networking managers to get more freedom in selecting networking hardware, Adaptable network architecture and robust protection of the network. These new trends are reviewed in this article. We are hopeful that this chapter provides future insights in the research of SDN.

DDoS and Cyber Attacks Detection and Mitigation in SDN: A Comprehensive Research of Moving Target Defense Systems
Neetu Gupta, Rupanshi Agarwal, Sukhvinder Singh Dari, Suraj Malik, Rahul Bhatt, and Dharmesh Dhabliya
IEEE
Software defined network (SDN) is an evolving networking technology. SDN supports high scalability and efficiency for end users. The involvement more devices or users to the SDN controller, security during data sharing is lacking. Especially, Distributed Denial of Service (DDoS) attackers target SDN controller and switches for the aim of exhaust the resources of them. To address those issues, authors in the current works have established various solutions such as firewall, machine learning and deep learning approaches to block the request from DDoS attackers and also addressed the issue of resource consumption and scalability. A Moving Target Defense (MTD) mechanism is one of the approaches for DDoS attacks detection and mitigation. MTD provides fault tolerance and security features. Different MTD schemes proposed such as Proxy Server assisted MTD, Flow Filtering MTD, and Manager aided MTD. Existing MTD suffers by many issues such as single point of failure, lack of traffic information, communication overhead. Theoretical analysis of various MTD mechanisms is taken into account for comparison. Among the several MTD, joint defense (proactive and reactive) aided MTD gives the better scalability, flexibility and adaptation. It is better under multi-controller SDN environment.

Recognition of Criminal Faces From Wild VideosSurveillance System Using VGG-16 Architecture
Rahul Bhatt, Suraj Malik, Rishabh Arora, Gaurav Agarwal, Shilpa Sharma, and Anishkumar Dhablia
IEEE
In this research, we strive toward creating a real-time structure for identifying people's faces in CCTV footage using methods from artificial intelligence as well as deep training. The standard CCTV apparatus is expensive & inadequate since it requires constant human supervision. Many groups, including law enforcement, may benefit from the automated identification method of features in CCTV pictures because of the minimal human interaction and low cost required. Scaling, rotation, noisy backdrops, & varying light intensities are only some of the challenges that plague image-based identification. By down sampling utilizing the VGG-16 construction, the suggested system is able to interpret pictures from surveillance footage in immediate time. Utilizing a facial tracker ID unit, it accurately identifies offenders & reduces forecast reversion by clearing up the crowded embed issue in the area of features and might arise during identity matches on a large number of face embedding databases. A recognition score building technique further improves the accuracy of the outcomes. In this research, we developed an operational version of the suggested system and tested the detection model, ultimately attaining a precision of 0.900 as well as an F-1 rating of 0.943. We additionally empirically validated that utilizing the monitored instance-level facial recognition approach described in this work, the frameworks suggested in earlier publications exhibit better performance. Rapid reactions like this have the potential to avert potential disasters, as well as the suggested system may be utilized to track down criminals as well as safeguard the nation's facilities.

A hexagonal topology and coordinate based routing for wmns
Suraj Malik, , Rakesh Kumar Singh, and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Wireless mesh networks (WMNs) contain both i.e mesh router and clients, where backbone of WMNs is mesh router due to immaterial portability. They give orchestrate access to both work and customary clients. In a WMN, it is feasible to accomplish an IGW or any Mesh Routers via various paths. Anyway this may prompt an unwanted circumstance, in which the most ideal ways may degenerate on account of the disbalancing of burden, subsequently bringing about problematic execution. To avoid this problem, we propose coordinate based multi-way routing protocol, which knows about area of its neighbors for an Infrastructural WMN. The principal thought is on the examination of directing estimation with speedy route revelation, as possible least hopes, least delays, most outrageous information rates and least possible defect-rates. Reproduction outcome, demonstrate that our introduced method estimation is preferable than current lattice, ETX and ETT metric employed in directing.

A novel block chain-based approach for secure handoff in WMN with reduced authentication delay

A secured and fault-tolerant multipath routing protocols for WMN
Paramjeet Rawat, Meenakshi Soam, and Suraj Malik
Springer Berlin Heidelberg

Dr. Suraj Malik

RESEARCH INTERESTS

Scopus Publications