Hierarchical Reinforcement Learning for Integrated Cloud-Fog-Edge Computing in IoT Systems

2025-11-12

AI Summary

To address high latency, poor scalability, and privacy risks in traditional cloud computing architectures when handling massive, time-sensitive IoT data, this paper proposes HIPA—a hierarchical cloud-fog-edge collaborative computing framework. HIPA introduces a layered heterogeneous processing architecture and innovatively employs hierarchical reinforcement learning (HRL) to enable dynamic task offloading and cross-layer resource coordination across edge, fog, and cloud tiers, supporting adaptive decision-making under dynamic environmental conditions. Compared with single-tier optimization approaches, HIPA achieves significant end-to-end latency reduction (38.2% average decrease in simulations) while preserving data locality, improves resource utilization and system throughput, and strengthens edge-side data privacy protection. Experimental results demonstrate HIPA’s effectiveness and scalability in latency-critical IoT applications.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Security and Privacy: Data transparency and provenance
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

The Internet of Things (IoT) is transforming industries by connecting billions of devices to collect, process, and share data. However, the massive data volumes and real-time demands of IoT applications strain traditional cloud computing architectures. This paper explores the complementary roles of cloud, fog, and edge computing in enhancing IoT performance, focusing on their ability to reduce latency, improve scalability, and ensure data privacy. We propose a novel framework, the Hierarchical IoT Processing Architecture (HIPA), which dynamically allocates computational tasks across cloud, fog, and edge layers using machine learning. By synthesizing current research and introducing HIPA, this paper highlights how these paradigms can create efficient, secure, and scalable IoT ecosystems.

Problem

• Optimizing IoT data processing across cloud, fog, and edge layers
• Reducing latency and improving scalability in IoT systems
• Dynamically allocating computational tasks using machine learning

Innovation

• Hierarchical Reinforcement Learning for IoT systems
• Dynamic task allocation across computing layers
• Machine learning optimizes cloud-fog-edge integration