Kamlesh Tiwari
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Kamlesh Tiwari has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
- Ashutosh Bhatia
Birla Institute of Technology and Science, Pilani
23 shared publications - Ankit Agrawal
Department of CSIS, Birla Institute of Technology and Science, Pilani Campus, Pilani, India
11 shared publications - Hari Mohan Pandey
Bournemouth University
11 shared publications - Santosh Kumar Yadav
University of Delhi
8 shared publications Phalguni Gupta
Indian Institute of Technology Kanpur
6 shared publicationsAchleshwar Luthra
Department of CSIS, Birla Institute of Technology and Science Pilani, Pilani Campus, Rajasthan 333031, India
3 shared publications
Latest Publications
- Open Access
- 20 November 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 164
Hierarchical Bayesian Deep Learning for return on advertising spend prediction: A probabilistic approach to e-commerce advertising
- 20 November 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 164
Open AccessHierarchical Bayesian Deep Learning for return on advertising spend prediction: A probabilistic approach to e-commerce advertising
- Open Access
- 30 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 6, 4923-4942
PUF-AQKD: A Hardware-Assisted Quantum Key Distribution Protocol for Man-in-the-Middle Attack Mitigation
The Quantum Key Distribution (QKD) protocol utilizes quantum mechanics principles for cryptographic key exchange, ensuring absolute secrecy. Current QKD techniques are susceptible to man-in-the-middle (MITM) attacks due to the absence of an inherent mechanism for identity verification within the quantum channel. For authentication, these systems rely on classical or post-quantum cryptography, which diminishes the perfect security advantage provided by QKD. We present a Physical Unclonable Function (PUF)-based authenticated QKD protocol (PUF-AQKD), which avoids the necessity for authenticated classical channels and is useful in mitigating MITM attacks. The fundamental concept of PUF-AQKD is to implement a phase shift in the basis used for polarizing the transmitted qubits. The phase shift is dictated by PUFs, which are anticipated to result in analogous (correlated) responses for devices manufactured under similar conditions but dissimilar responses in different conditions. An adversary lacking a correlated PUF response shared by Alice and Bob would inadvertently increase the Quantum Bit Error Rates (QBER) observed at Bob’s end. We present a mathematical model to assess the efficacy of the proposed PUF-AQKD method and perform simulations utilizing the NetSquid simulator. The mathematical analysis and simulation findings indicate that PUF-AQKD can efficiently identify eavesdroppers, even during incomplete measurements, without the necessity of an authorized classical channel.
- 30 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 6, 4923-4942
Open AccessPUF-AQKD: A Hardware-Assisted Quantum Key Distribution Protocol for Man-in-the-Middle Attack Mitigation
The Quantum Key Distribution (QKD) protocol utilizes quantum mechanics principles for cryptographic key exchange, ensuring absolute secrecy. Current QKD techniques are susceptible to man-in-the-middle (MITM) attacks due to the absence of an inherent mechanism for identity verification within the quantum channel. For authentication, these systems rely on classical or post-quantum cryptography, which diminishes the perfect security advantage provided by QKD. We present a Physical Unclonable Function (PUF)-based authenticated QKD protocol (PUF-AQKD), which avoids the necessity for authenticated classical channels and is useful in mitigating MITM attacks. The fundamental concept of PUF-AQKD is to implement a phase shift in the basis used for polarizing the transmitted qubits. The phase shift is dictated by PUFs, which are anticipated to result in analogous (correlated) responses for devices manufactured under similar conditions but dissimilar responses in different conditions. An adversary lacking a correlated PUF response shared by Alice and Bob would inadvertently increase the Quantum Bit Error Rates (QBER) observed at Bob’s end. We present a mathematical model to assess the efficacy of the proposed PUF-AQKD method and perform simulations utilizing the NetSquid simulator. The mathematical analysis and simulation findings indicate that PUF-AQKD can efficiently identify eavesdroppers, even during incomplete measurements, without the necessity of an authorized classical channel.
- Open Access
- 15 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 86413-86429
Bayesian Deep Learning Meets Self-Attention: A Risk-Aware Approach to Advertisement Optimization
In the highly competitive landscape of e-commerce advertising, maximizing Return on Advertising Spend (ROAS) is critical, yet remains inherently uncertain due to auction-based bidding dynamics and fluctuating market conditions. Traditional deterministic models fail to capture this uncertainty, necessitating a probabilistic approach that balances predictive accuracy with interpretability. To address this challenge, the paper proposes a novel Hierarchical Bayesian Deep Learning framework that integrates a Bayesian Belief Network (BBN) for structured probabilistic reasoning and a Mixture Density Network (MDN) for full distributional modeling of ROAS. The BBN models dependencies among campaign variables, offering interpretable insights, while the hierarchical deep learning architecture overcomes scalability limitations in high-dimensional settings through self-attention mechanisms. Experiments demonstrate up to 22.8% lower RMSE and 27.4% better Negative Log Likelihood (NLL) and up to 31.2% lower Kullback-Leibler divergence (KLD) than state-of-the-art methods (DeepAR, Prophet, NGBoost), achieving an R² of 98% with an inference speed of 5.2 ms per campaign, making real-time bidding feasible. Ablation studies confirm that attention-driven feature selection and calibrated uncertainty quantification significantly enhance both predictive performance and explainability, identifying key drivers of campaign success. By providing precise, uncertainty-aware, and explainable predictions, this approach enables adaptive bidding strategies, optimized budget allocation, and risk management, setting a new benchmark for intelligent decision-making in digital advertising.
- 15 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 86413-86429
Open AccessBayesian Deep Learning Meets Self-Attention: A Risk-Aware Approach to Advertisement Optimization
In the highly competitive landscape of e-commerce advertising, maximizing Return on Advertising Spend (ROAS) is critical, yet remains inherently uncertain due to auction-based bidding dynamics and fluctuating market conditions. Traditional deterministic models fail to capture this uncertainty, necessitating a probabilistic approach that balances predictive accuracy with interpretability. To address this challenge, the paper proposes a novel Hierarchical Bayesian Deep Learning framework that integrates a Bayesian Belief Network (BBN) for structured probabilistic reasoning and a Mixture Density Network (MDN) for full distributional modeling of ROAS. The BBN models dependencies among campaign variables, offering interpretable insights, while the hierarchical deep learning architecture overcomes scalability limitations in high-dimensional settings through self-attention mechanisms. Experiments demonstrate up to 22.8% lower RMSE and 27.4% better Negative Log Likelihood (NLL) and up to 31.2% lower Kullback-Leibler divergence (KLD) than state-of-the-art methods (DeepAR, Prophet, NGBoost), achieving an R² of 98% with an inference speed of 5.2 ms per campaign, making real-time bidding feasible. Ablation studies confirm that attention-driven feature selection and calibrated uncertainty quantification significantly enhance both predictive performance and explainability, identifying key drivers of campaign success. By providing precise, uncertainty-aware, and explainable predictions, this approach enables adaptive bidding strategies, optimized budget allocation, and risk management, setting a new benchmark for intelligent decision-making in digital advertising.
- Open Access
- 29 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Computer Society
- Vol. 6, 586-598
Sybil-Resilient Publisher Selection Mechanism in Blockchain-Based MCS Systems
In Blockchain-based Mobile CrowdSensing (BMCS) systems, publishers (data collectors) can exploit the ability to create multiple blockchain identities, enabling Sybil attacks. Selfish, malicious, and collusive Sybil behaviors undermine both reward and majority-based data validation mechanisms, discouraging honest participation and threatening system integrity. Existing solutions often fail to address these issues, particularly in environments dominated by selfish or malicious publishers. This paper proposes a novel two-phase publisher selection mechanism to mitigate Sybil attacks in BMCS systems. Phase-I employs a modified Proof-of-Stake (PoS) mechanism with carefully calibrated parameters, including staked amount, coinage, reputation, and randomness. The strategic combination of staked amount and coinage increases the difficulty of Sybil attacks as the system scales over time. Phase-II introduces a lightweight, reputation-based Proof-of-Work (PoW) mechanism tailored for Mobile CrowdSensing (MCS) environments, where puzzle difficulty adjusts dynamically based on the publisher's reputation. Reputation and penalization mechanisms are central to the proposed mechanism, ensuring robust prevention of task domination, selfish behavior, and malicious activities while fostering honest participation. Comprehensive on-chain and off-chain simulations demonstrate the proposed mechanism's effectiveness in mitigating Sybil attacks, reducing their impact, and promoting fair participation.
- 29 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Computer Society
- Vol. 6, 586-598
Open AccessSybil-Resilient Publisher Selection Mechanism in Blockchain-Based MCS Systems
In Blockchain-based Mobile CrowdSensing (BMCS) systems, publishers (data collectors) can exploit the ability to create multiple blockchain identities, enabling Sybil attacks. Selfish, malicious, and collusive Sybil behaviors undermine both reward and majority-based data validation mechanisms, discouraging honest participation and threatening system integrity. Existing solutions often fail to address these issues, particularly in environments dominated by selfish or malicious publishers. This paper proposes a novel two-phase publisher selection mechanism to mitigate Sybil attacks in BMCS systems. Phase-I employs a modified Proof-of-Stake (PoS) mechanism with carefully calibrated parameters, including staked amount, coinage, reputation, and randomness. The strategic combination of staked amount and coinage increases the difficulty of Sybil attacks as the system scales over time. Phase-II introduces a lightweight, reputation-based Proof-of-Work (PoW) mechanism tailored for Mobile CrowdSensing (MCS) environments, where puzzle difficulty adjusts dynamically based on the publisher's reputation. Reputation and penalization mechanisms are central to the proposed mechanism, ensuring robust prevention of task domination, selfish behavior, and malicious activities while fostering honest participation. Comprehensive on-chain and off-chain simulations demonstrate the proposed mechanism's effectiveness in mitigating Sybil attacks, reducing their impact, and promoting fair participation.
- 22 April 2025
- Published by Springer Nature
- p. 257-269
A Layered Framework for Blockchain Security: Classification of Threats and the Quantum Computing Impact
- 22 April 2025
- Published by Springer Nature
- p. 257-269
A Layered Framework for Blockchain Security: Classification of Threats and the Quantum Computing Impact
- 22 April 2025
- Published by Springer Nature
- p. 281-292
Layered Blockchain-Based Mobile CrowdSensing Architecture: Exploring Privacy and Scalability Challenges Across Layers
- 22 April 2025
- Published by Springer Nature
- p. 281-292
Layered Blockchain-Based Mobile CrowdSensing Architecture: Exploring Privacy and Scalability Challenges Across Layers
- 18 April 2025
- Published by Springer Nature
- p. 378-389
Decentralized Marketplace for Maintenance of Electric Vehicles
- 18 April 2025
- Published by Springer Nature
- p. 378-389
Decentralized Marketplace for Maintenance of Electric Vehicles
- 15 April 2025
- Published by Springer Nature
- p. 152-164
ExProCO: An Explainable Probabilistic Campaign Optimizer for eCommerce Advertising
- 15 April 2025
- Published by Springer Nature
- p. 152-164
ExProCO: An Explainable Probabilistic Campaign Optimizer for eCommerce Advertising
- 8 April 2025
- Published by Springer Nature
- p. 415-427
TrPrNet: Early Parkinson Detection Network Using Marker-Less Gait Analysis
- 8 April 2025
- Published by Springer Nature
- p. 415-427
TrPrNet: Early Parkinson Detection Network Using Marker-Less Gait Analysis
- 15 March 2025
- Published by Springer Nature
- p. 99-107
A Touchless Palm-Photo Recognition System for Mobile and Handheld Devices
- 15 March 2025
- Published by Springer Nature
- p. 99-107
A Touchless Palm-Photo Recognition System for Mobile and Handheld Devices