B. N. V. Madhu Babu
B. N. V. Madhu Babu has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Burri Naresh
Department of CSE-CS, CVR College of Engineering, Hyderabad, India
1 shared publication - C. S. Sowjanya
Department of CSE Sreyas Institute of Engineering and Technology, Hyderabad, India
1 shared publication - Ch Naga Santhosh Kumar
JNTU
1 shared publication Hanumantha Rao Battu
Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Andhra Pradesh, India
1 shared publication- K. Venkataramana
CSE (DS) Department, Malla Reddy Engineering College, Telangana, India
1 shared publication - Ravikumar Ch
Chaitanya Bharathi Institute of Technology
1 shared publication
Latest Publications
- Open Access
- 28 March 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 178
An optimized blockchain framework with ML-based anomaly detection for distributed access control in IoT networks
The fast rate of adoption of cloud-integrated Internet of Things (IoT) environs poses critical problems to access control because of the resource restrictiveness, heterogeneity, and dynamism of IoT devices, making the traditional centralized security models not applicable. To overcome these shortcomings, this paper presents a blockchain-based distributed access control system CapBlock, which combines the ability-based access control (CapBAC), smart contracts, and machine learning (ML) to provide dynamic, scalable, and resistant security enforcement. The framework applies the Ensemble Cuckoo Search Optimization (ECSO) algorithm, Bloom filters and the public-key cryptography to optimize the usage of resources, reduce the latency associated with authorization and improve the scalability of the system, and the use of a Random Forest-based anomaly detection model is continuously used to monitor the behavior of users to access the system and dynamically update the access policies to prevent malicious activities. The system is tested on a large scale of simulation with 4,000 access transactions on metrics such as accuracy, precision, recall, F1-score, transaction latency, throughput, security of data and preservation of privacy. The user privacy protection rate is considered as the percentage of requests of access where sensitive user attributes are not disclosed during authentication, authorization, and blockchain transaction logging, where the average value is 93 percent. Moreover, the throughput is calculated by scaling loads of request-per-second in HTTP requests to blockchain transactions commits with the assumption of one validated request per one completed transaction, under which the proposed framework can reach the highest throughput of 1,200 transactions per second (TPS) with stable performance. The experimental findings support the notion that CapBlock is a highly secure system capable of promoting security resiliency, privacy protection, and scalable access control in the contemporary cloud-IoT systems.
- 28 March 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 178
Open AccessAn optimized blockchain framework with ML-based anomaly detection for distributed access control in IoT networks
The fast rate of adoption of cloud-integrated Internet of Things (IoT) environs poses critical problems to access control because of the resource restrictiveness, heterogeneity, and dynamism of IoT devices, making the traditional centralized security models not applicable. To overcome these shortcomings, this paper presents a blockchain-based distributed access control system CapBlock, which combines the ability-based access control (CapBAC), smart contracts, and machine learning (ML) to provide dynamic, scalable, and resistant security enforcement. The framework applies the Ensemble Cuckoo Search Optimization (ECSO) algorithm, Bloom filters and the public-key cryptography to optimize the usage of resources, reduce the latency associated with authorization and improve the scalability of the system, and the use of a Random Forest-based anomaly detection model is continuously used to monitor the behavior of users to access the system and dynamically update the access policies to prevent malicious activities. The system is tested on a large scale of simulation with 4,000 access transactions on metrics such as accuracy, precision, recall, F1-score, transaction latency, throughput, security of data and preservation of privacy. The user privacy protection rate is considered as the percentage of requests of access where sensitive user attributes are not disclosed during authentication, authorization, and blockchain transaction logging, where the average value is 93 percent. Moreover, the throughput is calculated by scaling loads of request-per-second in HTTP requests to blockchain transactions commits with the assumption of one validated request per one completed transaction, under which the proposed framework can reach the highest throughput of 1,200 transactions per second (TPS) with stable performance. The experimental findings support the notion that CapBlock is a highly secure system capable of promoting security resiliency, privacy protection, and scalable access control in the contemporary cloud-IoT systems.
- Open Access
- 2 August 2025
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 15 (4) , 25532-25537
A Hybrid Approach for Fraud Detection in Digital Wallet Transactions Using Adversarial Autoencoders and Gated Recurrent Units
Digital payment systems are increasingly used to complete everyday transactions; however, their digital nature exposes users to the risk of fraudulent activity, necessitating advanced detection techniques to ensure security. This study proposes a hybrid fraud detection model for digital wallet transactions by integrating Adversarial Autoencoders (AAE) and Gated Recurrent Units (GRU), combining AAE's ability to learns robust latent representations, and GRU's ability to capture temporal dependencies within transaction sequences. The proposed method outperforms existing approaches, achieving 99% accuracy, 99% recall, 98% F1-score, 99% precision, and 99.4% Area Under the Curve (AUC). By effectively reducing both false positives and false negatives, the model improves fraud detection and mitigates financial risks in digital transactions.
- 2 August 2025
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 15 (4) , 25532-25537
Open AccessA Hybrid Approach for Fraud Detection in Digital Wallet Transactions Using Adversarial Autoencoders and Gated Recurrent Units
Digital payment systems are increasingly used to complete everyday transactions; however, their digital nature exposes users to the risk of fraudulent activity, necessitating advanced detection techniques to ensure security. This study proposes a hybrid fraud detection model for digital wallet transactions by integrating Adversarial Autoencoders (AAE) and Gated Recurrent Units (GRU), combining AAE's ability to learns robust latent representations, and GRU's ability to capture temporal dependencies within transaction sequences. The proposed method outperforms existing approaches, achieving 99% accuracy, 99% recall, 98% F1-score, 99% precision, and 99.4% Area Under the Curve (AUC). By effectively reducing both false positives and false negatives, the model improves fraud detection and mitigates financial risks in digital transactions.