The Energy-Aware Hierarchical Green Fog framework is presented, which introduces a unified reinforcement learning (RL) orchestration layer that explicitly incorporates residual energy, renewable energy availability, spatial proximity, and task deadlines into hierarchical fog-cloud decision-making.
Abstract
Industrial Internet of Things (IIoT) systems face growing demands for low-latency, energy-efficient, and trustworthy operation under heterogeneous devices, mobility, and renewable energy variability. Existing fog-cloud approaches typically optimize isolated objectives and lack integrated mechanisms for sustainability and verifiable coordination. This paper presents the Energy-Aware Hierarchical Green Fog (EAHGF) framework, which introduces a unified reinforcement learning (RL) orchestration layer that explicitly incorporates residual energy, renewable energy availability, spatial proximity (via BLE), and task deadlines into hierarchical fog-cloud decision-making. A lightweight Proof-of-Stake blockchain provides immutable auditability of allocations with minimal overhead. A stochastic multi-layer queuing model captures system dynamics, while RL-based scheduling and proximity-aware offloading jointly optimize energy and latency. Extensive OMNeT++/INET simulations with up to 3,000 heterogeneous IIoT devices (Poisson arrivals λ = 0.5-2 tasks/s, random waypoint mobility 1-5 m/s, 70% renewable offset on fog nodes) demonstrate that EAHGF achieves a workload acceptance rate of ~ 92%, reduces energy consumption by approximately 28%, and improves latency by ~ 22% compared to baseline fog frameworks and FogNetSim++. The integrated PoS blockchain maintains ~ 100 ms confirmation latency while providing blockchain-assisted accountability, traceability, and trust in resource allocation decisions. EAHGF thus offers a scalable, sustainable, and trustworthy foundation for next-generation Green IIoT deployments, preserving ~ 65% residual energy versus ~ 45% in conventional systems.
The ubiquitous deployment of Internet of Things (IoT) in smart building ecosystems generates massive volumes of multi-dimensional data, rendering secure storage and efficient retrieval paramount challenges. Although blockchain technology ensures data integrity and traceability, applying it to resource-constrained IoT networks exposes a fundamental “storage trilemma” among cost, latency, and scalability. Conventional approaches, relying on either static local retention or full cloud offloading, fail to reconcile these conflicting objectives. In this paper, we propose a Heat-Driven Hybrid Storage (HDHS) architecture that addresses limitations of existing hybrid storage systems-which rely on static parameters and reactive policiesthrough three key innovations: predictive heat modeling, dynamic redundancy adaptation, and multi-objective optimization. Specifically, HDHS incorporates a time-decay model with cost-aware uncertainty estimation to forecast block access “heat” under noisy conditions. Based on these predictions, the system dynamically tunes redundancy rates and utilizes rateless fountain codes to optimize the trade-off between storage footprint and data durability. Furthermore, we design a cloud-window optimizer that addresses a multi-objective trade-off to determine an effective boundary for local-cloud data migration. Extensive experiments on real-world datasets demonstrate that our scheme achieves a 40.7% reduction in storage costs, maintains sub-3ms query latency for 74.7% of queries, and ensures 99% + data reliability in permissioned blockchain environments.
Wei Yang, Xiaohua Wu, Yichang Chen et al.· Annual International Compute...· 0 citations
Industrial Internet of Things (IIoT) applications in smart manufacturing generate dynamic and latency-sensitive workloads that challenge conventional fog-cloud resource allocation methods. To address this problem, this paper proposes SDRA-IIoT, a predictive and multi-objective orchestration framework for smart manufacturing applications deployed across IIoT, fog, and cloud layers. The framework integrates surrogate workload estimation, MOPSO-based placement optimization, security-aware filtering, QoS-aware validation, and adaptive load balancing within a unified runtime allocation process. SDRA-IIoT is implemented in iFogSim and evaluated against Hybrid Fuzzy-DQN and IoT DRL Offloading under a common simulation setting using repeated-run analysis over five random seeds. Experimental results show that SDRA-IIoT achieves the best overall service-level performance, reducing makespan to $164.43 \pm 13.99 \mathrm{~ms}$, lowering average response latency to $334.18 \pm 160.87 \mathrm{~ms}$, improving resource utilization efficiency to $89.02 \pm 0.55 \%$, and decreasing the deadline miss rate to 8.35±4.02%. These results demonstrate the effectiveness of predictive and multi-objective fog-cloud orchestration for industrial resource management.
E. Nagarjun, Dharamendra Chouhan· 2026 5th International Confe...· 0 citations
Smart city operations increasingly depend on timely and secure data management, yet Fog-IoT ecosystems face persistent issues related to latency, energy consumption, and data integrity. This study proposes a Cloud-Fog-IoT architecture integrated with a Directed Acyclic Graph (DAG)-based blockchain to enhance scalability and security. Experimental evaluation was conducted using a 10-node peer-to-peer topology (8:2 transaction mix of reads to writes, 1 MB block size, 500 ms block timeout) across Raspberry Pi 4B devices (4GB RAM, gigabit Ethernet). Compared to a cloud-IoT baseline (centralized AWS m5.large, mean network latency 70 ms) and a fog-only configuration (Intel NUC edge nodes, direct aggregation), the proposed system demonstrated a mean transaction throughput of 560 tx/s (95% CI: ), compared with 430 tx/s (95% CI: ) and 350 tx/s (95% CI: ), respectively. Mean end-to-end latency was reduced to 220 ms (95% CI: ), compared to 290 ms and 310 ms for cloud-IoT and fog-only systems. Energy consumption decreased to 0.22 Joules/tx (95% CI: [0.21, 0.23]), versus 0.37 and 0.30 Joules/tx for the baselines. The architecture supported 15 concurrent nodes versus 9 for the cloud-IoT system and achieved a transaction success rate of 99% (cloud-IoT: 90%; fog-only: 91%). Results indicate robust and scalable performance for secure smart city deployments, with consistent definitions and units throughout the manuscript.
Path planning for cloud based autonomous systems such as smart transportation, Internet of Things (IoT) installations and robot fleets need to be secure, energy efficient, time efficient and fulfil privacy constraints. Current reinforcement learning (RL) techniques mainly consider optimisation of single objective, centralised or loosely secured model updates which are susceptible to data poisoning, privacy breach and adversarial model updates. We present Blockchain enabled Energy and Time efficient Multi Objective Reinforcement Learning (BlockE2T MORL) a new decentralised approach for secure, cloud assisted path planning. BlockE2T MORL has three main components: (i) a dynamic multi objective reward function that reduces energy, travel time and security threat; (ii) a lightweight blockchain inspired trust mechanism that assigns continuous trust values to agents, and is incorporated in the reward function to punish dishonest or malicious agents; and (iii) a hybrid actor critic learning strategy that facilitates exploration and exploitation in dynamic environments. Unlike conventional blockchain systems, our approach incurs low computational overhead ([Formula: see text] instead of [Formula: see text] for validation) and operates without heavy consensus protocols. We evaluate BlockE2T-MORL on a simulated grid-based cloud environment with up to 100 agents and varying adversarial ratios. After 200 training episodes (5 independent runs), the proposed framework achieves: (i) energy consumption = 124.3 ± 8.7 J (23.4% reduction vs. standard RL, p < 0.01), (ii) latency = 45.2 ± 3.8 ms (17.4% improvement, p < 0.05), and (iii) trust score = 0.87 ± 0.04 (67% improvement, p < 0.001). The framework converges faster (210 ± 25 episodes in the final optimized configuration, compared with ≥ 520 episodes for baseline methods). BlockE2T-MORL offers a scalable, privacy-preserving, and computationally lightweight solution for next-generation intelligent path planning in cloud-based autonomous systems.
Revati Raman Dewangan, D. Thombre, Vivek Parganiha et al.· Scientific Reports· 0 citations
Amid the explosive growth of latency-aware and computation-sensitive services, mobile edge computing (MEC) assisted by aerial networks, such as high-altitude platforms (HAPs) and low-altitude unmanned aerial vehicles (UAVs), has emerged as an effective solution for providing computational capabilities to regions with sparse terrestrial infrastructure. Nevertheless, aerial networks are highly sensitive to energy cost and inherently constrained in hosting dense computing resources, while the exposed wireless environment renders them particularly vulnerable to attacks from malicious nodes. Consequently, it is imperative to develop effective task scheduling and computing resources management mechanisms that satisfy users quality-of-service (QoS) requirements while minimizing system cost and ensuring network reliability. In this paper, we develop a multi-cell MEC network composed of multiple UAVs and multiple HAPs, and further propose a dual-layer blockchain-enabled, federated election (FE)-based (DBFE) group relative policy optimization (GRPO) algorithm to jointly reduce the task offloading latency and system energy expenditure. In particular, blockchain techniques enhance system resilience against non-Byzantine failures, whereas the FE mechanism suppresses the influence of Byzantine behaviors. Simulation results demonstrate that, compared with existing approaches, the proposed method reduces the overall system cost by 19% and 31% under scenarios without malicious nodes and with malicious nodes, respectively.
Haoyu Wan, Meng Li, Qi Li et al.· IEEE Transactions on Cogniti...· 0 citations
IoT cloud platforms support the monitoring and management of distributed renewable-energy systems, but sustaining them can be challenging in academic and community initiatives with limited resources. Recurring cloud-service fees, dependence on proprietary platforms, and limited local control may become major obstacles, particularly when long-term service operation must be combined with technical capacity-building within local institutions and surrounding communities. This paper presents an open-source, multi-tenant IoT cloud architecture developed within the GREATER Erasmus+ framework and deployed in renewable-energy Living Labs in Rwanda. The architecture is built using widely adopted open-source tools, including Docker, Node-RED, MySQL, Nginx, MQTT, and HTTPS APIs. The main novelty is an open-source middleware layer that transforms standard components into a shared, tenant-aware platform by coordinating authentication, role-based access, tenant-aware routing, controlled database access, separate Node-RED workspaces, and dashboard visibility for different Living Labs and user roles. The platform has been deployed in real settings, supporting photovoltaic monitoring, solar-powered irrigation, community energy services, domestic energy monitoring, and educational activities. The platform is evaluated in terms of cost, resource usage, communication delay, reliability mechanisms, and access-control behavior. The results show that heterogeneous IoT services can be hosted on modest local infrastructure while maintaining logical separation between Living Lab data and services. By combining open-source technologies, multi-tenant management, and field deployment, the proposed architecture offers a practical model for higher-education institutions and community-oriented renewable-energy initiatives in resource-constrained environments.