Carbon-neutral Cybersecurity Circuits Using Osv-bisga for Sustainable Industrial Applications
Year: 2026
Type: article
Abstract: Industrial cyber–physical systems increasingly have relied on secure circuits that also have aligned with carbon neutral goals. Prior studies have emphasized detection accuracy, yet the energy footprint of security algorithms has remained marginally addressed. The need for energy-aware cybersecurity circuits has therefore emerged as a critical research direction. Conventional intrusion detection circuits have consumed excessive power due to continuous monitoring that has depended on computationally intensive learning models. These approaches have limited suitability for green industrial infrastructure, where both security assurance and energy efficiency have been demanded simultaneously. A lack of unified frameworks that have integrated low-power algorithms with sustainable circuit deployment has persisted. A novel energy-efficient framework that has been termed OSV-BiSGA has been proposed for industrial cybersecurity circuits. The framework has combined a one-class support vector model with a bidirectional snow geese optimization algorithm that has minimized redundant computations. Continuous monitoring that has been designed at the circuit level has adapted sampling rates dynamically, which has reduced idle power consumption. Green infrastructure principles that have included low-leakage components and adaptive voltage scaling have been incorporated into the circuit design. Optimization that has guided parameter selection has ensured minimal energy usage without degrading detection reliability. The proposed OSV-BiSGA framework achieves an accuracy of 0.95, precision of 0.93, recall of 0.94, and F1-score of 0.94 at a population size of 30, while reducing energy consumption to 40 J. Compared with Static One Class SVM, PSO-Optimized IDS, and Always-On Deep Learning IDS, the framework reduces energy usage by up to 63% while maintaining superior detection performance.
Author: Thomas Samraj Lawrence
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7127630508