• Categorization of existing service orchestration strategies that improve energy-efficiency in Edge computing. • Identification and analysis of novel approaches introduced by integrating renewable energy sources into Edge service orchestration. • Open research challenges and case studies to future sustainable orchestration. Edge computing is an emerging paradigm of decentralized computing. It extends the Cloud infrastructure to the Edge of the network and closer to users. It allows the deployment of low-latency services, better privacy and extended availability. Orchestration, based on softwarization and virtualization of services, plays a crucial role in performance optimization and dynamic adaptation to the evolving context. It constitutes the basis of the operation of Edge computing. In this systematic literature review we study the problem of energy efficiency in the orchestration of Edge services. Orchestration represents both a challenge and a lever for the energy management of future systems of this type that will be deployed on a global scale. Faced with current environmental concerns and the growing and plethora of heterogeneous and scattered resources involved in these systems, it is imperative to review the design of orchestration methods from the additional perspective of sustainability and efficiency. We provide a comprehensive classification framework specifically designed for energy-efficient Edge service orchestration, analyzing 63 research papers across traditional and renewable energy contexts. Our analysis reveals three distinct categories of traditional strategies and identifies three emerging orchestration paradigms unique to renewable energy integration. We explore the research landscape, identifying strategies that consider renewable energy, energy storage and renewable energy management, lacking in the current survey literature. We also discuss eleven potential future directions and provide three case studies. This review strengthens the critical importance of addressing sustainability concerns and demonstrates how virtualization and service orchestration constitute fundamental foundations for future energy-efficient distributed ICT infrastructures.

Federated Class-Incremental Learning faces critical challenges including catastrophic forgetting, semantic drift, and privacy risks under non-IID data distributions. To address these, we propose FedCapD, a novel framework that unifies unsupervised task boundary detection via Bayesian nonparametric modeling, hierarchical semantic distillation through capsule alignment and GNN-based class propagation, secure gradient communication using homomorphic encryption and differential privacy, and diffusion-based generative replay for memory-efficient adaptation. By integrating structural reasoning with privacy-preserving learning, FedCapD achieves state-of-the-art performance across four diverse datasets-CheXpert, MIMIC-CXR-JPG, BraTS2021, and PHM2012-in terms of semantic consistency, encrypted distillation fidelity, cold-start accuracy, and utility efficiency under privacy constraints. The framework eliminates reliance on real-data storage and supports scalable, lifelong learning in regulated, resource-constrained environments such as healthcare AI and edge computing. Code is available at FCIL .

Microwave drying (MWD) is a promising technique for dehydrating agricultural products due to its rapid volumetric heating, high energy efficiency, and superior preservation of product quality. However, the complex and nonlinear nature of microwave–material interactions, along with the spatial and temporal variability of dielectric properties, presents significant challenges for process modeling, control, and optimization. Traditional mathematical models often fall short in capturing these dynamics, which limits their use in adaptive or real-time process regulation. The goal of this review is to provide a comprehensive synthesis of artificial intelligence (AI) techniques applied to the microwave drying of agricultural products, focusing on predictive modeling, intelligent control, and optimization through a bibliometric analysis that covers literature from 2014 to 2024. Techniques such as artificial neural networks (ANNs), support vector machines (SVMs), adaptive neuro-fuzzy inference systems (ANFIS), and evolutionary algorithms are assessed for their effectiveness in modeling drying kinetics, predicting quality attributes, and supporting closed-loop control. Recent advancements in hybrid and ensemble models, real-time sensor integration, and multi-objective optimization are also examined. The review highlights current limitations in AI-based drying systems, including data scarcity, overfitting, poor model interpretability, and limited real-time deployment. It proposes strategic future directions, such as the adoption of explainable AI, digital twin frameworks, embedded edge computing, and sensor fusion for autonomous control. This work highlights the transformative potential of AI in developing intelligent, scalable, and energy-efficient MWD systems that align with the goals of Industry 4.0 and sustainable food engineering. • AI methods predict microwave drying behavior more effectively than classical models. • ANNs are widely used to model drying rates and moisture ratios from microwave parameters. • Hybrid AI models improve optimization of microwave power and temperature settings. • AI-driven control systems use sensor feedback and machine learning to adjust drying. • AI enables multi-objective optimization of energy use, time, and product quality.

Digital twin is emerging as a key technology to monitor the status of complex industry systems. Valuable insights, such as running statuses and anomalies, can be analyzed from the collected system status timely. Considering that the data updating from each system component (known as a physical object) to its digital twin is performed continuously, timely and accurate streaming analytics based on machine learning models is a key technology to analyze such data efficiently. In this paper, we focus on enabling low-delay yet highly-accurate streaming analytics for digital twin applications in mobile edge computing (MEC) networks. Specifically, we formulate a fundamental optimization problem of digital twin placements and model selections for streaming analytics, with the aim of minimizing both the analytic loss and the processing delay. To this end, we first consider the problem with a single query, for which, we propose an approximation algorithm with provable approximation ratio for a special case, and then devise an efficient algorithm for the original problem with a single query. We then study the online digital twin placement and model selection problem for streaming analytics with multiple queries under real scenarios, where resource demands of arrival queries and resource availability of MEC network are uncertain. We propose an online learning algorithm with a bounded regret to make admission policies. We finally evaluate the performance of the proposed algorithms by extensive simulations. Results show that the weighted sums of the total processing delay and the cumulative loss in the solution delivered by the proposed algorithms outperform their counterparts by 12.5% with a single query and 13.3% with multiple queries, respectively.

Enhancing computer education through IoT-Enabled learning environments leveraging mobile edge computing for real-time feedback

The underlying challenges in the wearable electronic market are the limited power and processing capability often resulting in failure to handle complex computations. In order to balance the computational demands with resource constraints, the potential of intelligent reflecting surfaces (IRSs) are exploited in this paper. The paper presents a wearable electronics network in which the wearable nodes communicate with the assistance of double faced active (DFA) IRSs. By simultaneously controlling reflection and transmission links with active amplification, DFA-IRS enables reliable mobile edge computing (MEC)-based task offloading from wearable devices to nearby processing nodes. A resource utilization (RU) algorithm is proposed that associates the devices with DFA-IRSs. The optimal phase shifts of DFA-IRSs are obtained. Further, the impact of transmit power [Formula: see text], number of DFA-IRSs [Formula: see text], number of DFA-IRS elements N, per element amplification [Formula: see text], power budget [Formula: see text] on the average sum rate of the system is evaluated. It is observed that the DFA-IRS aided system offers average sum rate of 8.2bps/Hz with N =120 and [Formula: see text] of 20dBm with optimal phase shifts [Formula: see text] and [Formula: see text] in the reflection and transmission space respectively. Also, there is an improvement of 5.80% in average sum rate over random phase shifts. The comparison with conventional IRS, single faced active (SFA)-IRS and simultaneously transmitting and reflecting (STAR) IRS is also presented. In the end, the use case of proposed network for personalized healthcare is discussed.

The constraints of current intelligent healthcare systems are extensively addressed in this study, which offers a thorough framework for expanding eHealth within the human-centered paradigm of Society 5.0. Reliance on centralised cloud systems, which have delay for real-time IoMT data, single points of failure, and serious data privacy/security threats, are some of the disadvantages of current approaches. Traditional healthcare AI models are often black-boxes, lacking explainability and transparency (XAI), which are necessary for patient acceptance and physician trust. To overcome these issues, the Proposed Approach uses a novel combination of specialised technologies and a multi-tiered architecture of cloud services, edge computing, and IoMT. The Health Prediction using Cloud Edge 2.0 (HPCE 2.0) algorithm employs fuzzy logic to combine static Electronic Health Records (EHRs) with dynamic real-time IoMT data to handle medical input uncertainty and imprecision. This algorithm predicts health severity accurately and individually. Proof of Authentication 2.0 (PoAh 2.0) consensus ensures data integrity and non-repudiation in the blockchain-enhanced architecture's immutable, decentralised ledger. By adding XAI (LIME/SHAP) to give local and alternative answers, the system stops being a black box and becomes an open collaborator. Edge-cloud integration improves performance by lowering delay for important real-time alerts. Security tests show that the PoAh 2.0 system works well and can be expanded by quickly building and verifying blocks. This makes sure that strict privacy rules are followed while keeping the good predictive performance of a cardiac arrest prediction case study. This platform sets a new standard for interpretable, safe, and responsive AI-driven healthcare.

Optimizing smart healthcare systems via integrated edge, fog, and cloud computing

The proposed Smart Surveillance System presents a novel, hardware-integrated prototype demonstration aimed The proposed Smart Surveillance System presents a groundbreaking hardware-integrated prototype that decisively validates the effectiveness of a dual-branch anti-spoofing model on the low-power edge device, Raspberry Pi 3B+. This prototype goes beyond algorithmic performance research, showcasing a fully functional proof of concept. In contrast to existing surveillance solutions that typically rely on centralized cloud processing or basic recognition systems, our system employs an advanced dual-branch model that utilizes both spatial and frequency-domain features. This approach enables real-time anti-spoofing with an impressive error rate of less than 2%. What truly sets our system apart is its seamless end-to-end integration of cloud-based authentication, edge-level inference for rapid response, and an interactive live video conferencing feature. This configuration empowers immediate verification and action during potential spoofing events. With IoT-enabled devices, our system ensures effortless communication for live streaming, automated alerts, and scalable cloud data management. Coupled with edge computing, it guarantees real-time decision-making with minimal latency. Experimental results confirm its high accuracy in distinguishing genuine users from spoofing attempts, positioning our solution as a lightweight, proactive, and user-interactive surveillance option that is perfectly suited for homes, enterprises, and public infrastructures.

Nowadays, farmers across the globe are gradually adopting intelligent farming, which is facilitated by a variety of cutting-edge technologies. The advancement of intelligent farming applications is greatly aided by the internet of farming things (IoFT). Massive IoFT devices generally possess constrained resources, making it challenging to meet the battery and computational requirements of intelligent farming applications through local computation. RF energy harvesting enabled mobile edge computing (RFE-MEC) addresses this issue by harvesting RF energy from an access point, offloading and computing tasks at the edge in a nearby access point. In the proposed scheme, multiuser nonorthogonal multiple access allows the IoFT devices to simultaneously offload computationally intensive tasks to the MEC server for processing. The delay outage probability closed-form expression is formulated for the RFE-NOMA-MEC intelligent farming system under a Rayleigh fading channel. The impact of imperfect channel state information on the RFE-NOMA-MEC is considered. Tunicate enhanced northern goshawk optimization algorithm (TNGO) has been proposed to discover the optimal parameter set to minimize delay outage probability. The results indicate that the system performance is enhanced using TNGO when the optimal time switching factor, power allocation coefficient and task allocation ratio are utilized.

This paper introduces a novel IoT-Enhanced Virtual Power Plant (VPP) framework that integrates edge-fog computing, blockchain-secured communication, and AI-driven market mechanisms to optimize energy management in smart grids. The proposed system addresses critical challenges in traditional VPPs including high latency (> 500ms), cybersecurity vulnerabilities (68% of grids report IoT risks), and low user engagement (< 30% adoption). Our framework achieves sub-50ms response times through hybrid edge-fog computing, resolves 94% of security vulnerabilities via blockchain consensus, and increases user participation by 116% through personalized demand response. Extensive simulations using real-world datasets from Pecan Street and prototype deployment with 120 IoT-connected distributed energy resources (DERs) demonstrate 23.1% improvement in energy efficiency, 17.3% cost reduction for end-users, and 142.3% return on investment over five years. The system integrates vehicle-to-grid (V2G) technology, achieving 25% better renewable energy utilization while maintaining grid stability. Implementation using open-source platforms (OpenEMS, GridLAB-D) ensures scalability up to 10,000 + DERs with modular architecture supporting community-driven innovation.

Abstract Vehicular Networks (VNs) are playing a key role in shaping the future of Intelligent Transportation Systems (ITS) by enabling real-time communication among vehicles (V2V), infrastructure (V2I), and broader entities (V2X). These networks offer immense potential for improving road safety, easing traffic congestion, and supporting smart mobility services. At the same time, their open and dynamic nature brings significant challenges related to privacy and security, including risks like unauthorized access, data breaches, location tracking, and denial-of-service attacks. This survey offers a comprehensive overview of the major security and privacy concerns confronted by the contemporary vehicular networks. It introduces a clear taxonomy that organizes various threat categories, defense strategies, and cryptographic techniques that are commonly used in this domain. Recent developments are reviewed and grouped based on various authentication schemes, blockchain-based methods, privacy-preserving protocols, and AI-driven intrusion detection systems. These solutions are then evaluated in terms of their cryptographic robustness, privacy safeguards, processing requirements, and their suitability for fitting into real-world applications such as autonomous driving and vehicular edge computing. Beyond the current landscape, the paper also outlines key gaps and promising research directions, including federated learning, lightweight blockchain consensus mechanisms, quantum-resistant cryptography, and intelligent threat detection at the edge. By bringing together both foundational concepts and the latest innovations, this survey aims to support researchers and practitioners in understanding the contemporary security challenges inherent to the vehicular networks and building more secure, scalable, and privacy-aware vehicular communication systems.

Abstract In this work, we present HiAER-Spike, a modular, reconfigurable, event-driven neuromorphic computing platform designed to execute large spiking neural networks with up to 160 million neurons and 40 billion synapses - roughly twice the neurons of a mouse brain at faster than real time. This system, assembled at the UC San Diego Supercomputer Center, comprises a co-designed hard- and software stack that is optimized for run-time massively parallel processing and hierarchical address-event routing (HiAER) of spikes while promoting memory-efficient network storage and execution. The architecture efficiently handles both sparse connectivity and sparse activity for robust and low-latency event-driven inference for both edge and cloud computing. A Python programming interface to HiAER-Spike, agnostic to hardware-level detail, shields the user from complexity in the configuration and execution of general spiking neural networks with minimal constraints in topology. The system is made easily available over a web portal for use by the wider community. In the following, we provide an overview of the hard- and software stack, explain the underlying design principles, demonstrate some of the system’s capabilities, and solicit feedback from the broader neuromorphic community. Examples are shown demonstrating HiAER-Spike’s capabilities for event-driven vision on benchmark CIFAR-10, DVS event-based gesture, MNIST, and Pong tasks.

The rapid deployment of 5G networks and the proliferation of Internet of Things (IoT) devices have significantly increased the complexity of urban electromagnetic radiation (EMR) environments. Conventional ground-based monitoring systems are spatially limited and unable to provide three-dimensional field characterization. This paper proposes an integrated IoT–UAV framework for high-resolution EMR monitoring, spatial reconstruction, and intelligent source classification. A four-layer architecture combining distributed sensing, edge computing, cloud analytics, and visualization is developed. A formal electromagnetic propagation model is introduced to ensure consistency between broadband exposure measurements and frequency-selective spectral analysis. A CNN–LSTM architecture is implemented for spectral–temporal source classification, achieving 95% validation accuracy across five EMR categories. Simulation-based validation demonstrates up to an eightfold improvement in spatial coverage compared to fixed ground networks while maintaining a practical anomaly detection threshold of −55 dBm in the spectrum-analysis RF chain. The proposed framework establishes a mathematically consistent and practically deployable solution for next-generation EMR monitoring systems.

Ultra-high-speed railway communication systems face several technical challenges due to extremely high mobility, including Doppler-induced channel variations, frequent handovers, and increasing network traffic. These challenges not only degrade communication reliability but also negatively affect the efficiency of network resource utilization. In this paper, we review the key technical challenges in ultra-high-speed railway communication environments and investigate artificial intelligence (AI)-based intelligent network control techniques to address these issues. In particular, we examine mobility management approaches focusing on AI-based predictive handover schemes and intelligent network control architectures based on the Open Radio Access Network (O-RAN). In addition, network resource management strategies are discussed through mobile edge computing (MEC)-enabled traffic offloading and task migration techniques. Through this analysis, we discuss the potential applicability of intelligent network control technologies for improving communication reliability and enhancing network resource utilization efficiency in ultra-high-speed railway communication environments.

• A data-driven framework merges IoT, edge, and local analytics for EV fleets. • Machine learning forecasts battery health to guide energy-aware operations. • Hybrid BiLSTM–GRU and TabNet–TCN models enhance accuracy and scalability. • The system improves energy efficiency and stabilizes maintenance planning. • It bridges technical forecasting with strategic fleet-management insights. Efficient energy and asset management is a key economic driver in the large-scale deployment of electric vehicles (EVs). This research develops a data‑driven management framework that integrates Internet-of-Things (IoT) connectivity, edge computing, and on-premises analytics to forecast battery state-of-health (SOH) and optimize lifecycle performance in EV fleets. The proposed multilayered architecture establishes a closed decision-support loop combining localized intelligence for rapid diagnostics with centralized analytics for long-term strategic optimization. Machine-learning models—including BiLSTM–GRU and TabNet–TCN hybrids—are employed within an operational context to balance prediction accuracy, computational cost, and system scalability. Validation using the NASA PCoE lithium-ion dataset confirms that this integrated approach enhances energy-use efficiency and reduces maintenance cost variability under real-world uncertainty. In addition, connecting battery health forecasting to broader economic considerations reinforces effective management strategies, supporting cost–efficient and sustainable decision–making in real EV operations. By linking technical forecasting with managerial decision insights, the framework supports sustainable fleet operation, strengthens predictive maintenance planning, and aligns with UN SDG 7 objectives on affordable and clean energy. This study therefore bridges the technical and managerial perspectives of energy conversion and management, demonstrating how intelligent analytics can inform effective decision-making in EV-based energy ecosystems.

The integration of machine learning, edge computing, and Internet of Things technologies is enabling real-time, energy-efficient, and context-aware monitoring in smart agriculture. This paper presents a scalable AIoT framework that combines dual communication protocols, LoRa and NB-IoT, with lightweight machine learning models deployed on resource-constrained embedded platforms. The system adaptively assigns workloads between edge and cloud based on data criticality and network conditions, ensuring low-latency inference and reliable operation. Hardware-aware benchmarking demonstrates that real-time prediction is feasible on ultra-low-power devices, achieving up to 86% accuracy with Decision Tree and Random Forest models, while maintaining low memory footprint and energy consumption. Field experiments show that edge inference consistently reduces latency (average 347–383 ms) and variability compared to cloud modes, and the hybrid communication design mitigates the impact of network contention and harsh wireless conditions. Security evaluations indicate that A Denial-of-Service (DoS), Replay, and Man-in-the-Middle (MITM) attacks are effectively mitigated with minimal overhead. These results highlight the framework’s practical applicability, energy efficiency, and resilience, providing a hardware-agnostic roadmap for deploying AIoT-enabled smart farming systems in diverse agricultural settings.

The scalable deployment of Edge AI within the Internet of Intelligent Things (IoIT) is currently limited by a disconnect between theoretical model performance and physical hardware realities. As a result, many real-world systems fail to account for practical constraints that can impair performance, robustness, and long-term operation of embedded AI implementations. To bridge this gap, this article proposes an assessment framework designed to quantify the operational viability of the Bare Edge (CPU-only Single-Board Computers) through four well-defined stages, aimed at the assessment of three competing constraints: Energy ( C 1 ), Cost ( C 2 ), and Performance ( C 3 ). We validate this methodology through an exhaustive characterisation of 400 unique hardware-software configurations across the Raspberry Pi ecosystem, using state-of-the-art YOLO11 object detection as the target workload. The application of this framework uncovers a counter-intuitive Efficiency Inversion, where high-performance architectures consume less energy per inference task than their predecessors, and rigorously quantifies a “Python Tax” that dictates thermal stability in passive environments. Synthesising these empirical findings, we formulate the Efficiency-Viability Matrix, a strategic decision-making tool that model three operational archetypes: The Sentinel for autonomy, The Live Tracker for speed, and The Analyst for precision. This study provides system architects with a scalable roadmap for exploiting the Bare Edge, demonstrating that widespread visual intelligence is viable without the reliance on dedicated hardware accelerators. Ultimately, this research provides the scientific and academic community with a reproducible baseline that bridges the critical gap between theoretical AI algorithms and their physical viability, fostering sustainable democratisation of intelligent systems.

Designing resilient IoT and Edge Computing with federated tinyML

Edge caching alleviates backhaul pressure and enhances video service quality by deploying video content near user devices. However, the limited storage capacity of edge servers struggles to cope with the exponential growth of video data, challenging the delivery of high-quality video services. While both Device-to-Device (D2D) caching and multi-bitrate video technology are promising solutions to relieve the pressure on edge servers, existing research suffers from a key limitation: studies on multi-bitrate caching are predominantly focused on the edge layer, while D2D caching is often limited to single-bitrate scenarios. This isolation neglects the significant benefits of integrating D2D caching with multi-bitrate technology and fails to develop a cross-layer caching strategy for multi-bitrate videos. To address this limitation, we propose a D2D-enhanced Multi-bitrate video cAching straTEgy (MATE) for cloud-edge-device collaborative networks. We formulate a joint service latency and caching replacement cost optimization problem, which can be modeled as a mixed-integer programming problem. To overcome the coupling between caching strategies at the edge layer and device layer, we employ an alternating iterative optimization approach to decouple the original problem into two subproblems. We design an edge-device double-layer joint caching strategy, i.e., a device-layer caching strategy based on greedy algorithm and Lagrange multipliers, and an edge-layer caching strategy based on multi-agent twin delayed deep deterministic policy gradient algorithm. Extensive simulations are conducted to demonstrate the effectiveness of the proposed MATE.