Edge Caching in Fog-Based Sensor Networks through Deep Learning-Associated Quantum Computing Framework

Area of Science:

• Computer Science
• Information Technology
• Quantum Computing

Background:

• Mobile edge computing (MEC) and fog computing (FC) offer edge services but face performance degradation and quality of service (QoS) issues in Internet of Things (IoT) scenarios.
• Existing caching strategies struggle to optimize limited storage and accelerate content delivery, necessitating enhanced approaches.

Purpose of the Study:

• To propose a novel framework merging deep learning (DL) with quantum computing (QC) to improve caching efficiency in fog computing (FC) based sensor networks.
• To enhance the cache hit ratio and accelerate content delivery in mobile edge computing (MEC) environments.

Main Methods:

• A deep learning (DL) agent utilizing Self-Organizing Maps (SOMs) algorithm prioritizes caching contents based on trending data.
• Prioritized content is stored in a Quantum Memory Module (QMM) leveraging the Two-Level Spin Quantum Phenomenon (TLSQP).
• Quantum parallelism (QP) principles of superposition and entanglement are utilized for efficient resource utilization and content storage.

Main Results:

• The SOMs algorithm identified high and medium-priority content with a low topographic and quantization error (0.0000235) after 750,000 iterations.
• The integration of DL and QC is shown to improve cache hit ratio by ranking and efficiently storing prioritized content.
• Reduced resource utilization compared to conventional computing is achieved through quantum parallelism.

Conclusions:

• The proposed DL-MEC framework with QMM effectively addresses performance deterioration and QoS degradation in IoT environments.
• This hybrid approach significantly enhances content delivery speed and optimizes storage management in edge computing.
• The study demonstrates the potential of quantum computing (QC) to revolutionize caching strategies in next-generation wireless networks.