Purushottam Singh

@bitmesra.ac.in

Computer Science and Engineering
Birla Institute of Technology Mesra

Purushottam Singh


                  Download Compact PDF  
https://researchid.co/psingh21

EDUCATION

M.Tech in Computer Science and Engineering
Ph.D in Network Security and Cryptography

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Engineering, Computer Networks and Communications, Artificial Intelligence, Computer Science
10

Scopus Publications

61

Scholar Citations

4

Scholar h-index

2

Scholar i10-index

Scopus Publications

  • A GA-GAN approach for next-generation cryptographic security with a focus on quantum-resistant cryptography
    Purushottam Singh, Prashant Pranav, Sandip Dutta
    Discover Computing, 2025
  • Optimizing cryptographic protocols against side channel attacks using WGAN-GP and genetic algorithms
    Purushottam Singh, Prashant Pranav, Sandip Dutta
    Scientific Reports, 2025
  • Bi-GAN-LDA for cybersecurity: a hybrid deep learning framework for advanced network anomaly detection
    Purushottam Singh, Prashant Pranav, Sandip Dutta
    Engineering Research Express, 2025
    Intrusion Detection Systems (IDS) play a crucial role in modern cybersecurity by identifying and mitigating malicious activities in network traffic. However, existing IDS models suffer from high false positive rates, class imbalance issues, and inefficient feature selection, which hinder their ability to detect sophisticated cyber threats. In this study, study proposes Bi-GAN-LDA IDS, a novel hybrid deep learning framework that integrates Bidirectional Generative Adversarial Networks (Bi-GANs) for synthetic attack sample generation and Linear Discriminant Analysis (LDA) for optimized feature selection. Additionally, a custom focal loss function is introduced to enhance the classification of minority attack classes. The efficacy of the proposed Bi-GAN-LDA intrusion detection framework was rigorously validated using a diverse set of benchmark datasets, namely NSL-KDD, UNSW-NB-15, CICIDS-2017, ADFA-LD, and UNR-IDD. Notably, on the ADFA-LD dataset, the model achieved an F1-score of 99.5%, marking a 2.8% performance gain over existing GAN-based IDS frameworks. Furthermore, a substantial 22% reduction in false positive rates was observed when compared to conventional deep learning-based detectors. These improvements underscore the robustness of the proposed method, particularly in addressing the challenge of class imbalance, minimizing false alarms, and enhancing the reliability of real-time anomaly detection in contemporary IDS environments.
  • Network Security and Cryptography: Threats, Obstacles and Solutions-A Bibliometric Analysis
    Purushottam Singh, Sandip Dutta, Prashant Pranav
    Recent Advances in Computer Science and Communications, 2025
    Background: In the wake of escalating cyber threats and the indispensability of robust network security mechanisms, it becomes crucial to understand the evolving landscape of cryptographic research. Recognizing the significant contributions and discerning emerging trends can guide future strategies and technological advancements. Our study endeavors to shed light on this through a bibliometric analysis of publications in the realms of Network Security and Cryptography. Method: To chronicle and synthesize the progression of research methodologies from their inception to the present day, we undertook a comprehensive Bibliometric Analysis of Network Security and Cryptography. Our data set was culled from the Clarivate Analytics Web of Science Database, encompassing 3,897 papers, 603 sources, and 7,886 authors from across the globe. Results: Our analysis revealed a marked upsurge in cryptographic research since 1992, with China standing out as a dominant contributor in terms of publications. Notably, while 'security' and 'cryptography' emerged as recurrent research themes, there's an observable downtrend in international collaborations. Our study also highlights pivotal topics shaping the network security domain, offering insights into the trajectories of research source growth, structural variabilities in research relevance, and prospective intellectual and collaborative avenues as guided by authorship patterns. Conclusion: Cryptographic research is on an upward trajectory, both in volume and significance. However, the tapering of international collaborations and an evident need to concentrate on emergent challenges, such as data privacy and innovative network attacks, emerge as notable insights. This bibliometric review serves as a compass, directing researchers and academicians towards areas warranting heightened attention, thereby informing the roadmap for future investigative pursuits.
  • Premier Dynamic Bandwidth Management and Tensile Wavelength Selection Ensuring QoS for NG-EPONs
    Purushottam Singh, Pushpendra Kumar, Harshita Patel, Kanojia Sindhuben Babulal, Gayathri Ananthakrishnan
    IEEE Access, 2025
    Next Generation Ethernet Passive Optical Networks (NG-EPONs) have emerged as a leading choice for global network connectivity due to their cost-effectiveness, enhanced security, and energy efficiency. As data demands continue to surge with technological advancements, the need for efficient bandwidth management and wavelength selection in NG-EPONs becomes paramount. This paper presents two innovative algorithms—Tensile Wavelength and Dynamic Bandwidth Allocation (TW-DBA) and Premier Dynamic Bandwidth Allocation (PDBA)—designed to optimize bandwidth allocation and maintain Quality of Service (QoS) under varying network conditions. The TW-DBA algorithm achieves a remarkable throughput of 2.34 Gbps, driven by its dynamic wavelength selection mechanism that accounts for factors such as ONU-OLT distance, power availability, and bandwidth demand. Comparative analysis reveals that TW-DBA outperforms existing algorithms like Flexible Wavelength (FW), First-Fit, and Water Filled, both in computational efficiency and resource allocation. Furthermore, the PDBA algorithm demonstrates a minimum blockage probability of 0.0025 for limited ONU scenarios and 0.005 in unlimited scenarios, ensuring uniform bandwidth distribution. Simulation results confirm the superior performance and effectiveness of the proposed models, positioning them as robust solutions for the evolving demands of NG-EPONs.
  • Anomaly Detection in IoT Networks Using WGAN-GP: A Novel Approach for Robust IoT Security
    Purushottam Singh, Prashant Pranav, Sandip Dutta, Prasunn Dubey, P. Parimalam
    Lecture Notes in Networks and Systems, 2025
  • Leveraging generative adversarial networks for enhanced cryptographic key generation
    Purushottam Singh, Prashant Pranav, Shamama Anwar, Sandip Dutta
    Concurrency and Computation Practice and Experience, 2024
    SummaryIn this research, we present an innovative cryptographic key generation method utilizing a Generative Adversarial Network (GAN), enhanced by Merkel tree verification, marking a significant advancement in cryptographic security. Our approach successfully generates a large 6272‐bit key, rigorously tested for randomness and reliability using the Dieharder and NIST test suites. This groundbreaking method harmoniously blends cutting‐edge machine learning techniques with traditional cryptographic verification, setting a new standard in data encryption and security. Our findings not only demonstrate the efficacy of GANs in producing highly secure cryptographic keys but also highlight the effectiveness of Merkel tree verification in ensuring the integrity of these keys. The integration of merkel tree in our method provides a means to efficiently verify the authenticity of the large generated key sets. This research has broad implications for the future of secure communications, providing a robust solution in a world increasingly reliant on digital security. The integration of machine learning and cryptographic principles opens up new avenues for research and development, promising to bolster security measures in an era where digital threats are constantly evolving. This work contributes significantly to the field of cryptography, offering a novel perspective and robust solutions to the challenges of digital data protection.
  • Prevention of sleep deprivation attack in MANET using cumulative priority based cluster head selection
    Ankita Kumari, Purushottam Singh, Prashant Pranav, Sandip Dutta, Soubhik Chakraborty
    Concurrency and Computation Practice and Experience, 2024
    SummaryIn the rapidly evolving domain of Mobile Ad‐hoc Networks (MANETs), where their deployment spans critical military operations to essential organizational communication infrastructures, the pervasive threat of security breaches casts a long shadow on the networks' operational integrity and reliability. Central among these threats are sleep deprivation attacks, a particularly insidious form of cyber aggression that exploits the inherent decentralized and self‐organizing characteristics of MANETs to exhaust the energy reserves of nodes, compromising the network's stability and performance. This paper embarks on a journey to confront this challenge head‐on, introducing a pioneering and holistic defense mechanism that integrates a cumulative priority‐based model for the selection of cluster heads, ingeniously augmented by the principles of Chebyshev's Inequality for optimal load balancing. This novel strategy is designed not only to counteract the direct impacts of sleep deprivation attacks but also to address the underlying vulnerabilities of MANETs that these attacks exploit. Through a rigorous series of simulations, conducted across a spectrum of network scenarios to test the resilience and adaptability of our proposed model, we have observed a commendable success rate of 98% in neutralizing sleep deprivation attacks. By leveraging the dynamic nature of MANETs and integrating advanced statistical methods for load distribution and cluster management, our model offers a robust framework that significantly improves network performance and energy efficiency. This, in turn, fosters a more sustainable and reliable network environment, crucial for the high‐stakes applications MANETs support. By championing a comprehensive and adaptable approach to security, this study promises to reinstate user trust and ensure the continued reliability of these indispensable networks, securing their place as a cornerstone of modern communication infrastructure in the face of evolving cyber threats.
  • Optimizing GANs for Cryptography: The Role and Impact of Activation Functions in Neural Layers Assessing the Cryptographic Strength
    Purushottam Singh, Sandip Dutta, Prashant Pranav
    Applied Sciences Switzerland, 2024
    Generative Adversarial Networks (GANs) have surfaced as a transformative approach in the domain of cryptography, introducing a novel paradigm where two neural networks, the generator (akin to Alice) and the discriminator (akin to Bob), are pitted against each other in a cryptographic setting. A third network, representing Eve, attempts to decipher the encrypted information. The efficacy of this encryption–decryption process is deeply intertwined with the choice of activation functions employed within these networks. This study conducted a comparative analysis of four widely used activation functions within a standardized GAN framework. Our recent explorations underscore the superior performance achieved when utilizing the Rectified Linear Unit (ReLU) in the hidden layers combined with the Sigmoid activation function in the output layer. The non-linear nature introduced by the ReLU provides a sophisticated encryption pattern, rendering the deciphering process for Eve intricate. Simultaneously, the Sigmoid function in the output layer guarantees that the encrypted and decrypted messages are confined within a consistent range, facilitating a straightforward comparison with original messages. The amalgamation of these activation functions not only bolsters the encryption strength but also ensures the fidelity of the decrypted messages. These findings not only shed light on the optimal design considerations for GAN-based cryptographic systems but also underscore the potential of investigating hybrid activation functions for enhanced system optimization. In our exploration of cryptographic strength and training efficiency using various activation functions, we discovered that the “ReLU and Sigmoid” combination significantly outperforms the others, demonstrating superior security and a markedly efficient mean training time of 16.51 s per 2000 steps. This highlights the enduring effectiveness of established methodologies in cryptographic applications. This paper elucidates the implications of these choices, advocating for their adoption in GAN-based cryptographic models, given the superior results they yield in ensuring security and accuracy.
  • GAN Cryptography
    Purushottam Singh, Prashant Pranav, Sandip Dutta
    Machine Learning in Healthcare and Security Advances Obstacles and Solutions, 2024

RECENT SCHOLAR PUBLICATIONS

  • Bi-GAN-LDA for cybersecurity: A hybrid deep learning framework for advanced network anomaly detection
    P Singh, P Pranav, S Dutta
    Engineering Research Express 7 (2), 025238 , 2025
    2025
    Citations: 4
  • A GA-GAN approach for next-generation cryptographic security with a focus on quantum-resistant cryptography
    P Singh, P Pranav, S Dutta
    Discover Computing 28 (1), 82 , 2025
    2025
    Citations: 3
  • Premier dynamic bandwidth management and tensile wavelength selection ensuring qos for ng-epons
    P Singh, P Kumar, H Patel, KS Babulal, G Ananthakrishnan
    IEEE Access 13, 17068-17082 , 2025
    2025
    Citations: 5
  • Optimizing cryptographic protocols against side channel attacks using WGAN-GP and genetic algorithms
    P Singh, P Pranav, S Dutta
    Scientific Reports 15 (1), 2130 , 2025
    2025
    Citations: 14
  • Anomaly Detection in IoT Networks Using WGAN-GP: A Novel Approach for Robust IoT Security
    P Singh, P Pranav, S Dutta, P Dubey, P Parimalam
    International Conference on Network Security and Blockchain Technology, 247-260 , 2025
    2025
    Citations: 1
  • A modified RC‐4 cryptosystems to enhance security by using negative key schedule
    P Singh, S Dutta, P Pranav
    Security and Privacy 7 (6), e438 , 2024
    2024
    Citations: 3
  • Leveraging generative adversarial networks for enhanced cryptographic key generation
    P Singh, P Pranav, S Anwar, S Dutta
    Concurrency and Computation: Practice and Experience 36 (22), e8226 , 2024
    2024
    Citations: 6
  • Prevention of sleep deprivation attack in MANET using cumulative priority based cluster head selection
    A Kumari, P Singh, P Pranav, S Dutta, S Chakraborty
    Concurrency and Computation: Practice and Experience 36 (16), e8118 , 2024
    2024
  • Unmasking the Digital Illusion: A Comprehensive Bibliometric Analysis of Deepfake Detection Research
    P Singh, P Pranav, V Nath, S Dutta
    9th International Conference on Nanoelectronics, Computational Intelligence … , 2024
    2024
    Citations: 2
  • Optimizing GANs for cryptography: the role and impact of activation functions in neural layers assessing the cryptographic strength
    P Singh, S Dutta, P Pranav
    Applied Sciences 14 (6), 2379 , 2024
    2024
    Citations: 17
  • GAN cryptography
    P Singh, P Pranav, S Dutta
    Machine learning in healthcare and security, 184-194 , 2024
    2024
    Citations: 4
  • Network Security and Cryptography: Threats, Obstacles and Solutions - A Bibliometric Analysis
    P Singh, S Dutta, P Pranav
    Recent Advances in Computer Science and Communications 17 (DOI:10.2174 … , 2024
    2024
    Citations: 2

MOST CITED SCHOLAR PUBLICATIONS

  • Optimizing GANs for cryptography: the role and impact of activation functions in neural layers assessing the cryptographic strength
    P Singh, S Dutta, P Pranav
    Applied Sciences 14 (6), 2379 , 2024
    2024
    Citations: 17
  • Optimizing cryptographic protocols against side channel attacks using WGAN-GP and genetic algorithms
    P Singh, P Pranav, S Dutta
    Scientific Reports 15 (1), 2130 , 2025
    2025
    Citations: 14
  • Leveraging generative adversarial networks for enhanced cryptographic key generation
    P Singh, P Pranav, S Anwar, S Dutta
    Concurrency and Computation: Practice and Experience 36 (22), e8226 , 2024
    2024
    Citations: 6
  • Premier dynamic bandwidth management and tensile wavelength selection ensuring qos for ng-epons
    P Singh, P Kumar, H Patel, KS Babulal, G Ananthakrishnan
    IEEE Access 13, 17068-17082 , 2025
    2025
    Citations: 5
  • Bi-GAN-LDA for cybersecurity: A hybrid deep learning framework for advanced network anomaly detection
    P Singh, P Pranav, S Dutta
    Engineering Research Express 7 (2), 025238 , 2025
    2025
    Citations: 4
  • GAN cryptography
    P Singh, P Pranav, S Dutta
    Machine learning in healthcare and security, 184-194 , 2024
    2024
    Citations: 4
  • A GA-GAN approach for next-generation cryptographic security with a focus on quantum-resistant cryptography
    P Singh, P Pranav, S Dutta
    Discover Computing 28 (1), 82 , 2025
    2025
    Citations: 3
  • A modified RC‐4 cryptosystems to enhance security by using negative key schedule
    P Singh, S Dutta, P Pranav
    Security and Privacy 7 (6), e438 , 2024
    2024
    Citations: 3
  • Unmasking the Digital Illusion: A Comprehensive Bibliometric Analysis of Deepfake Detection Research
    P Singh, P Pranav, V Nath, S Dutta
    9th International Conference on Nanoelectronics, Computational Intelligence … , 2024
    2024
    Citations: 2
  • Network Security and Cryptography: Threats, Obstacles and Solutions - A Bibliometric Analysis
    P Singh, S Dutta, P Pranav
    Recent Advances in Computer Science and Communications 17 (DOI:10.2174 … , 2024
    2024
    Citations: 2
  • Anomaly Detection in IoT Networks Using WGAN-GP: A Novel Approach for Robust IoT Security
    P Singh, P Pranav, S Dutta, P Dubey, P Parimalam
    International Conference on Network Security and Blockchain Technology, 247-260 , 2025
    2025
    Citations: 1
  • Prevention of sleep deprivation attack in MANET using cumulative priority based cluster head selection
    A Kumari, P Singh, P Pranav, S Dutta, S Chakraborty
    Concurrency and Computation: Practice and Experience 36 (16), e8118 , 2024
    2024