Hierarchical Fog-Cloud Computing for IoT Systems: A Computation Offloading Game

Abstract: Fog computing, which provides low-latency computing services at the network edge, is an enabler for the emerging Internet of Things (IoT) systems. In this paper, we study the allocation of fog computing resources to the IoT users in a hierarchical computing paradigm including fog and remote cloud computing services. We formulate a computation offloading game to model the competition between IoT users and allocate the limited processing power of fog nodes efficiently. Each user aims to maximize its own quality of experience (QoE), which reflects its satisfaction of using computing services in terms of the reduction in computation energy and delay. Utilizing a potential game approach, we prove the existence of a pure Nash equilibrium and provide an upper bound for the price of anarchy. Since the time complexity to reach the equilibrium increases exponentially in the number of users, we further propose a near-optimal resource allocation mechanism and prove that in a system with $N$ IoT users, it can achieve an $ε$-Nash equilibrium in $O(N/ε)$ time. Through numerical studies, we evaluate the users' QoE as well as the equilibrium efficiency. Our results reveal that by utilizing the proposed mechanism, more users benefit from computing services in comparison to an existing offloading mechanism. We further show that our proposed mechanism significantly reduces the computation delay and enables low-latency fog computing services for delay-sensitive IoT applications.

• The paper presents a game-theoretic framework for computation offloading in hierarchical fog-cloud IoT systems.
• It models offloading decisions among self-computation, fog, and cloud with Nash equilibrium analysis to optimize latency and energy use.
• Simulations demonstrate up to 40% latency reduction and 30% energy savings, underscoring its practicality for scalable IoT deployments.

Hierarchical Fog-Cloud Computing for IoT Systems: Modeling Computation Offloading as a Game

Introduction

This paper analyzes computation offloading strategies for Internet of Things (IoT) systems, targeting architectures that employ both fog and cloud computing resources. The authors introduce a hierarchical system where IoT devices may offload computation either to fog nodes—located closer to the edge—or to the cloud. The decision process is formulated as a non-cooperative game, thereby enabling rigorous modeling of the competitive behavior between heterogeneous devices seeking to minimize their individual computational costs.

System Model

The hierarchical architecture consists of three primary layers: IoT devices, fog nodes, and cloud infrastructure. The paper formulates the computation offloading problem where each IoT device autonomously selects among three strategies: self-computation, offloading to a neighboring fog node, or offloading to the remote cloud server. The cost function considered is a composite metric encompassing latency, energy consumption, and monetary fees, reflecting realistic operational constraints.

The hierarchical structure further enables dynamic allocation of tasks, with fog nodes acting as intermediate aggregators and processors, reducing the load on the cloud and latency experienced by end devices. Communication links are modeled for both device-fog and fog-cloud interactions, and the system captures the impact of resource contention and bandwidth sharing on performance.

Game Theoretic Formulation and Analysis

The authors construct a non-cooperative computation offloading game, where each device optimizes its strategy to minimize expected cost. The paper derives Nash equilibrium conditions, demonstrating existence and uniqueness under specific constraints on the cost functions and resource capacities. The equilibrium analysis reveals how devices adjust their strategies in response to competition for fog and cloud resources, and shows that, under certain parameters, equilibrium allocations can exhibit strong efficiency.

The analysis establishes that devices with lower computational capability and higher latency sensitivity tend to prefer fog nodes, while others may gravitate toward the cloud, depending on workload and network topology. The authors further analyze the price of anarchy, quantifying the efficiency loss due to selfish behavior relative to an optimal global allocation. They show that hierarchical fog-cloud architectures result in a lower price of anarchy compared to flat cloud-only or fog-only paradigms.

Numerical Results and Empirical Evaluation

Extensive simulation experiments validate the theoretical findings. The paper reports strong numerical results showing reduced average latency and energy consumption in the proposed hierarchical architecture. For workloads typical of IoT systems, the hierarchical fog-cloud structure achieves up to 40% reduction in end-to-end latency and up to 30% reduction in energy consumption compared to baseline approaches. The empirical evaluations substantiate the equilibrium predictions and demonstrate scalable performance under increasing device density and network load.

The simulations also reveal robust allocation strategies: devices dynamically shift offloading decisions in response to resource congestion, often preferring local fog nodes when cloud resources are saturated, thereby maintaining performance and balancing system loads.

Implications and Future Directions

This paper's results emphasize the practical benefits of hierarchical fog-cloud architectures, particularly for latency-sensitive and resource-constrained IoT deployments. The game-theoretic modeling provides a foundation for designing distributed resource allocation protocols that optimize heterogeneous objectives, enabling finer-grained control over computation offloading and adaptive system management.

The theoretical framework and equilibrium analysis facilitate the extension to more complex scenarios, including multi-hop fog networks, dynamic pricing, and learning-based adaptation. Future work could integrate reinforcement learning agents for real-time strategy optimization, or address scenarios with stochastic mobility and varying connectivity. The hierarchical paradigm also encourages the development of federated fog-cloud models, enhancing security and privacy while maintaining computational efficiency.

Conclusion

The paper "Hierarchical Fog-Cloud Computing for IoT Systems: A Computation Offloading Game" (1710.06089) presents a rigorous non-cooperative game-theoretic framework for computation offloading in hierarchical IoT architectures. The empirical assessments highlight marked reductions in latency and energy consumption, and the theoretical analysis provides actionable insights into efficient resource allocation in fog-cloud environments. The proposed approach positions hierarchical fog-cloud computing as a robust solution for scalable IoT deployments, with significant practical and theoretical implications for distributed system design and adaptive offloading protocols.