Quantum readout and gradient deep learning model for secure and sustainable data access in IWSN

Area of Science:

• Computer Science
• Network Security
• Artificial Intelligence

Background:

• Industrial wireless sensor networks (IWSNs) face significant security and energy consumption challenges.
• Unmonitored environments in industrial plants increase vulnerability to unauthorized data access.
• Secure and energy-efficient authentication is crucial for sustainable IWSN data access.

Purpose of the Study:

• To propose a novel Quantum Readout Gradient Secured Deep Learning (QR-GSDL) model for IWSNs.
• To enhance security, energy efficiency, and green sustainability in industrial networks.
• To ensure only trustworthy sensors can access IWSN data.

Main Methods:

• A registration method using quantum readout and hash functions for efficient sensor onboarding.
• A gradient secured deep learning approach for energy-saving and secure authentication and authorization.
• Simulations comparing QR-GSDL against established anomaly detection and CNN models.

Main Results:

• The QR-GSDL model demonstrates high security and energy efficiency for IWSN applications.
• Experimental results show superior performance over existing models in key metrics.
• Key performance indicators include reduced energy consumption, higher authentication rates, faster authentication times, and lower false acceptance rates.

Conclusions:

• The QR-GSDL model effectively addresses security and energy efficiency challenges in IWSNs.
• The proposed model contributes to green sustainability by optimizing network resource utilization.
• QR-GSDL offers a promising solution for secure and sustainable data access in complex industrial environments.