Future 6G networks are envisioned to enhance the user experience in a multitude of different ways. The unification of existing terrestrial networks with non-terrestrial network (NTN) components will provide users with ubiquitous connectivity. Multi-access edge computing (MEC) will enable low-latency services, with computations performed closer to the end users, and distributed learning paradigms. Advanced multiple access schemes, such as sparse code multiple access (SCMA), can be employed to efficiently move data from edge nodes to spaceborne MEC servers. However, the non-orthogonal nature of SCMA results in interference, limiting the effectiveness of traditional SCMA receivers. Hence, NTN links should be protected with robust channel codes, significantly reducing the uplink throughput. Thus, we investigate the application of artificial intelligence (AI) to SCMA receivers for 6G NTNs. We train an AI model with multi-task learning to optimally separate and receive superimposed SCMA signals. Through link level simulations, we evaluate the block error rate (BLER) and the aggregated theoretical throughput achieved by the AI model as a function of the received energy per bit over noise power spectral density ratio (Eb/N0). We show that the proposed receiver achieves a target 10% BLER with 3.5dB lower Eb/N0 with respect to the benchmark algorithm. We conclude the assessment discussing the complexity-related challenges to the implementation of the AI model on board of a low earth orbit satellite.
Edge computing addresses the growing data demands of connected-device networks by placing computational resources closer to end users through decentralized infrastructures. This decentralization challenges traditional, fully centralized orchestration, which suffers from latency and resource bottlenecks. We present \textbf{FAuNO} -- \emph{Federated Asynchronous Network Orchestrator} -- a buffered, asynchronous \emph{federated reinforcement-learning} (FRL) framework for decentralized task offloading in edge systems. FAuNO adopts an actor-critic architecture in which local actors learn node-specific dynamics and peer interactions, while a federated critic aggregates experience across agents to encourage efficient cooperation and improve overall system performance. Experiments in the \emph{PeersimGym} environment show that FAuNO consistently matches or exceeds heuristic and federated multi-agent RL baselines in reducing task loss and latency, underscoring its adaptability to dynamic edge-computing scenarios.
As distributed energy resources (DERs) such as solar PV, batteries and electric vehicles become increasingly prevalent at the edge, maintaining grid stability requires advanced monitoring and control mechanisms. This paper presents a scalable smart grid gateway architecture that enables interoperability between Modbus-based inverters and IEEE 2030.5 cloud-based control systems. The proposed solution leverages Azure cloud services and edge-computing gateway devices to support dynamic configuration, telemetry ingestion, remote control and Volt-VAR Curve deployment. A microservice-based architecture ensures flexibility and scalability across diverse deployment scenarios, including both gateway-mediated and direct-to-cloud device communication. Results demonstrate the successful mapping of a Fronius Primo inverter's Modbus registers to IEEE 2030.5-compliant telemetry and control functions. Additionally, we evaluate real-time VVC updates and their impact on local voltage regulation, showcasing dynamic cloud-to-edge control with minimal latency. This work highlights the potential of virtualised, standards-based control infrastructures to support DER integration and active grid participation, while remaining adaptable to evolving smart grid architectures.
Accurate and robust pose estimation is a key requirement for any autonomous robot. We present cuVSLAM, a state-of-the-art solution for visual simultaneous localization and mapping, which can operate with a variety of visual-inertial sensor suites, including multiple RGB and depth cameras, and inertial measurement units. cuVSLAM supports operation with as few as one RGB camera to as many as 32 cameras, in arbitrary geometric configurations, thus supporting a wide range of robotic setups. cuVSLAM is specifically optimized using CUDA to deploy in real-time applications with minimal computational overhead on edge-computing devices such as the NVIDIA Jetson. We present the design and implementation of cuVSLAM, example use cases, and empirical results on several state-of-the-art benchmarks demonstrating the best-in-class performance of cuVSLAM.
Privacy-preserving distributed machine learning (ML) and aerial connected vehicle (ACV)-assisted edge computing have drawn significant attention lately. Since the onboard sensors of ACVs can capture new data as they move along their trajectories, the continual arrival of such 'newly' sensed data leads to online learning and demands carefully crafting the trajectories. Besides, as typical ACVs are inherently resource-constrained, computation- and communication-efficient ML solutions are needed. Therefore, we propose a computation- and communication-efficient online aerial federated learning (2CEOAFL) algorithm to take the benefits of continual sensed data and limited onboard resources of the ACVs. In particular, considering independently owned ACVs act as selfish data collectors, we first model their trajectories according to their respective time-varying data distributions. We then propose a 2CEOAFL algorithm that allows the flying ACVs to (a) prune the received dense ML model to make it shallow, (b) train the pruned model, and (c) probabilistically quantize and offload their trained accumulated gradients to the central server (CS). Our extensive simulation results show that the proposed 2CEOAFL algorithm delivers comparable performances to its non-pruned and nonquantized, hence, computation- and communication-inefficient counterparts.
Artificial intelligence (AI) has been increasingly applied to the condition monitoring of vehicular equipment, aiming to enhance maintenance strategies, reduce costs, and improve safety. Leveraging the edge computing paradigm, AI-based condition monitoring systems process vast streams of vehicular data to detect anomalies and optimize operational performance. In this work, we introduce a novel vehicle condition monitoring service that enables real-time diagnostics of a diverse set of anomalies while remaining practical for deployment in real-world edge environments. To address mobility challenges, we propose a closed-loop service orchestration framework where service migration across edge nodes is dynamically triggered by network-related metrics. Our approach has been implemented and tested in a real-world race circuit environment equipped with 5G network capabilities under diverse operational conditions. Experimental results demonstrate the effectiveness of our framework in ensuring low-latency AI inference and adaptive service placement, highlighting its potential for intelligent transportation and mobility applications.
Metaverse and Digital Twin (DT) have attracted much academic and industrial attraction to approach the future digital world. This paper introduces the advantages of deep reinforcement learning (DRL) in assisting Metaverse system-based Digital Twin. In this system, we assume that it includes several Metaverse User devices collecting data from the real world to transfer it into the virtual world, a Metaverse Virtual Access Point (MVAP) undertaking the processing of data, and an edge computing server that receives the offloading data from the MVAP. The proposed model works under a dynamic environment with various parameters changing over time. The experiment results show that our proposed DRL algorithm is suitable for offloading tasks to ensure the promptness of DT in a dynamic environment.
Mobile edge computing (MEC) allows appliances to offload workloads to neighboring MEC servers that have the potential for computation-intensive tasks with limited computational capabilities. This paper studied how deep reinforcement learning (DRL) algorithms are used in an MEC system to find feasible decentralized dynamic computation offloading strategies, which leads to the construction of an extensible MEC system that operates effectively with finite feedback. Even though the Deep Deterministic Policy Gradient (DDPG) algorithm, subject to their knowledge of the MEC system, can be used to allocate powers of both computation offloading and local execution, to learn a computation offloading policy for each user independently, we realized that this solution still has some inherent weaknesses. Hence, we introduced a new approach for this problem based on the Twin Delayed DDPG algorithm, which enables us to overcome this proneness and investigate cases where mobile users are portable. Numerical results showed that individual users can autonomously learn adequate policies through the proposed approach. Besides, the performance of the suggested solution exceeded the conventional DDPG-based power control strategy.
Physics-driven artificial intelligence (PD-AI) reconstruction methods have emerged as the state-of-the-art for accelerating MRI scans, enabling higher spatial and temporal resolutions. However, the high resolution of these scans generates massive data volumes, leading to challenges in transmission, storage, and real-time processing. This is particularly pronounced in functional MRI, where hundreds of volumetric acquisitions further exacerbate these demands. Edge computing with FPGAs presents a promising solution for enabling PD-AI reconstruction near the MRI sensors, reducing data transfer and storage bottlenecks. However, this requires optimization of PD-AI models for hardware efficiency through quantization and bypassing traditional FFT-based approaches, which can be a limitation due to their computational demands. In this work, we propose a novel PD-AI computational MRI approach optimized for FPGA-based edge computing devices, leveraging 8-bit complex data quantization and eliminating redundant FFT/IFFT operations. Our results show that this strategy improves computational efficiency while maintaining reconstruction quality comparable to conventional PD-AI methods, and outperforms standard clinical methods. Our approach presents an opportunity for high-resolution MRI reconstruction on resource-constrained devices, highlighting its potential for real-world deployment.
Wildfires continue to inflict devastating human, environmental, and economic losses globally, as tragically exemplified by the 2025 Los Angeles wildfire and the urgent demand for more effective response strategies. While physics-based and deep learning models have advanced wildfire simulation, they face critical limitations in predicting and visualizing multimodal fire spread in real time, particularly in both 2D and 3D spatial domains using dynamically updated GIS data. These limitations hinder timely emergency response, infrastructure protection, and community safety. Generative AI has recently emerged as a transformative approach across research and industry. Models such as Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), Transformers, and diffusion-based architectures offer distinct advantages over traditional methods, including the integration of multimodal data, generation of diverse scenarios under uncertainty, and improved modeling of wildfire dynamics across spatial and temporal scales. This position paper advocates for the adoption of generative AI as a foundational framework for wildfire prediction. We explore how such models can enhance 2D fire spread forecasting and enable more realistic, scalable 3D simulations. Additionally, we employ a novel human-AI collaboration framework using large language models (LLMs) for automated knowledge extraction, literature synthesis, and bibliometric mapping. Looking ahead, we identify five key visions for integrating generative AI into wildfire management: multimodal approaches, AI foundation models, conversational AI systems, edge-computing-based scenario generation, and cognitive digital twins. We also address three major challenges accompanying these opportunities and propose potential solutions to support their implementation.
This paper aims to develop an energy-efficient classifier for time-series data by introducing PatchEchoClassifier, a novel model that leverages a reservoir-based mechanism known as the Echo State Network (ESN). The model is designed for human activity recognition (HAR) using one-dimensional sensor signals and incorporates a tokenizer to extract patch-level representations. To train the model efficiently, we propose a knowledge distillation framework that transfers knowledge from a high-capacity MLP-Mixer teacher to the lightweight reservoir-based student model. Experimental evaluations on multiple HAR datasets demonstrate that our model achieves over 80 percent accuracy while significantly reducing computational cost. Notably, PatchEchoClassifier requires only about one-sixth of the floating point operations (FLOPS) compared to DeepConvLSTM, a widely used convolutional baseline. These results suggest that PatchEchoClassifier is a promising solution for real-time and energy-efficient human activity recognition in edge computing environments.
To achieve the Internet of Things (IoT) vision,Mobile Edge Computing (MEC) is a promising technology aimed at providing low-latency computing services to user equipment (UE). However, terrestrial MEC network struggles to provide service to UEs in remote and maritime region. Low Earth Orbit (LEO) satellite networks have the potential to overcome geographical restrictions and provide seamless global coverage for UEs. In this paper, we provide the first attempt to use stochastic geometry to investigate the performance of implementing space caching with LEO satellites (SATs) in the MEC network. We study a LEO satellite-assisted space caching MEC network, and LEO SATs can be equipped with servers to enable space caching, with the advantage of seamless coverage to assist terrestrial CSs for serving UEs in remote or maritime reigon. Using stochastic geometry and queuing theory, we establish an analytical framework for this MEC network. Meanwhile, we develop association strategies for UEs to connect with LEO SATs or CSs and utilize stochastic geometry to derive uplink and downlink coverage probabilities, considering the diversity of task and service types. On this basis, we employ the queuing theory to calculate the average delay to evaluate the system performance. Through Monte Carlo simulations and numerical results, the system performance is evaluated. The results show the potential of SAT spatial caching in improving the performance of the MEC network. Additionally, our results reveal useful insights such as the significant impact of the altitude and number of LEO SATs on the average delay of the network, providing helpful system-level recommendations for the design and configuration of the space-caching MEC network.
Foundation models like CLIP (Contrastive Language-Image Pretraining) have revolutionized vision-language tasks by enabling zero-shot and few-shot learning through cross-modal alignment. However, their computational complexity and large memory footprint make them unsuitable for deployment on resource-constrained edge devices, such as in-car cameras used for image collection and real-time processing. To address this challenge, we propose Clip4Retrofit, an efficient model distillation framework that enables real-time image labeling on edge devices. The framework is deployed on the Retrofit camera, a cost-effective edge device retrofitted into thousands of vehicles, despite strict limitations on compute performance and memory. Our approach distills the knowledge of the CLIP model into a lightweight student model, combining EfficientNet-B3 with multi-layer perceptron (MLP) projection heads to preserve cross-modal alignment while significantly reducing computational requirements. We demonstrate that our distilled model achieves a balance between efficiency and performance, making it ideal for deployment in real-world scenarios. Experimental results show that Clip4Retrofit can perform real-time image labeling and object identification on edge devices with limited resources, offering a practical solution for applications such as autonomous driving and retrofitting existing systems. This work bridges the gap between state-of-the-art vision-language models and their deployment in resource-constrained environments, paving the way for broader adoption of foundation models in edge computing.
With the advancement of edge computing, federated learning (FL) displays a bright promise as a privacy-preserving collaborative learning paradigm. However, one major challenge for FL is the data heterogeneity issue, which refers to the biased labeling preferences among multiple clients, negatively impacting convergence and model performance. Most previous FL methods attempt to tackle the data heterogeneity issue locally or globally, neglecting underlying class-wise structure information contained in each client. In this paper, we first study how data heterogeneity affects the divergence of the model and decompose it into local, global, and sampling drift sub-problems. To explore the potential of using intra-client class-wise structural knowledge in handling these drifts, we thus propose Federated Learning with Structural Knowledge Collaboration (FedSKC). The key idea of FedSKC is to extract and transfer domain preferences from inter-client data distributions, offering diverse class-relevant knowledge and a fair convergent signal. FedSKC comprises three components: i) local contrastive learning, to prevent weight divergence resulting from local training; ii) global discrepancy aggregation, which addresses the parameter deviation between the server and clients; iii) global period review, correcting for the sampling drift introduced by the server randomly selecting devices. We have theoretically analyzed FedSKC under non-convex objectives and empirically validated its superiority through extensive experimental results.
Neuromorphic computing, inspired by the human brain's neural architecture, is revolutionizing artificial intelligence and edge computing with its low-power, adaptive, and event-driven designs. However, these unique characteristics introduce novel cybersecurity risks. This paper proposes Neuromorphic Mimicry Attacks (NMAs), a groundbreaking class of threats that exploit the probabilistic and non-deterministic nature of neuromorphic chips to execute covert intrusions. By mimicking legitimate neural activity through techniques such as synaptic weight tampering and sensory input poisoning, NMAs evade traditional intrusion detection systems, posing risks to applications such as autonomous vehicles, smart medical implants, and IoT networks. This research develops a theoretical framework for NMAs, evaluates their impact using a simulated neuromorphic chip dataset, and proposes countermeasures, including neural-specific anomaly detection and secure synaptic learning protocols. The findings underscore the critical need for tailored cybersecurity measures to protect brain-inspired computing, offering a pioneering exploration of this emerging threat landscape.
The demand for machine learning (ML) model training on edge devices is escalating due to data privacy and personalized service needs. However, we observe that current on-device model training is hampered by the under-utilization of on-device data, due to low training throughput, limited storage and diverse data importance. To improve data resource utilization, we propose a two-stage data selection framework {\sf Titan} to select the most important data batch from streaming data for model training with guaranteed efficiency and effectiveness. Specifically, in the first stage, {\sf Titan} filters out a candidate dataset with potentially high importance in a coarse-grained manner.In the second stage of fine-grained selection, we propose a theoretically optimal data selection strategy to identify the data batch with the highest model performance improvement to current training round. To further enhance time-and-resource efficiency, {\sf Titan} leverages a pipeline to co-execute data selection and model training, and avoids resource conflicts by exploiting idle computing resources. We evaluate {\sf Titan} on real-world edge devices and three representative edge computing tasks with diverse models and data modalities. Empirical results demonstrate that {\sf Titan} achieves up to $43\%$ reduction in training time and $6.2\%$ increase in final accuracy with minor system overhead, such as data processing delay, memory footprint and energy consumption.
Integer quantization has emerged as a critical technique to facilitate deployment on resource-constrained devices. Although they do reduce the complexity of the learning models, their inference performance is often prone to quantization-induced errors. To this end, we introduce InTreeger: an end-to-end framework that takes a training dataset as input, and outputs an architecture-agnostic integer-only C implementation of tree-based machine learning model, without loss of precision. This framework enables anyone, even those without prior experience in machine learning, to generate a highly optimized integer-only classification model that can run on any hardware simply by providing an input dataset and target variable. We evaluated our generated implementations across three different architectures (ARM, x86, and RISC-V), resulting in significant improvements in inference latency. In addition, we show the energy efficiency compared to typical decision tree implementations that rely on floating-point arithmetic. The results underscore the advantages of integer-only inference, making it particularly suitable for energy- and area-constrained devices such as embedded systems and edge computing platforms, while also enabling the execution of decision trees on existing ultra-low power devices.
The development of edge computing places critical demands on energy-efficient model deployment for multiple-input multiple-output (MIMO) detection tasks. Deploying deep unfolding models such as PGD-Nets and ADMM-Nets into resource-constrained edge devices using quantization methods is challenging. Existing quantization methods based on quantization aware training (QAT) suffer from performance degradation due to their reliance on parametric distribution assumption of activations and static quantization step sizes. To address these challenges, this paper proposes a novel kernel-based adaptive quantization (KAQ) framework for deep unfolding networks. By utilizing a joint kernel density estimation (KDE) and maximum mean discrepancy (MMD) approach to align activation distributions between full-precision and quantized models, the need for prior distribution assumptions is eliminated. Additionally, a dynamic step size updating method is introduced to adjust the quantization step size based on the channel conditions of wireless networks. Extensive simulations demonstrate that the accuracy of proposed KAQ framework outperforms traditional methods and successfully reduces the model's inference latency.
Contemporary models of high dimensional physical systems are constrained by the curse of dimensionality and a reliance on dense data. We introduce KHRONOS (Kernel Expansion Hierarchy for Reduced Order, Neural Optimized Surrogates), an AI framework for model based, model free and model inversion tasks. KHRONOS constructs continuously differentiable target fields with a hierarchical composition of per-dimension kernel expansions, which are tensorized into modes and then superposed. We evaluate KHRONOS on a canonical 2D, Poisson equation benchmark: across 16 to 512 degrees of freedom (DoFs), it obtained L_2-square errors of 5e-4 down to 6e-11. This represents a greater than 100-fold gain over Kolmogorov Arnold Networks (which itself reports a 100 times improvement on MLPs/PINNs with 100 times fewer parameters) when controlling for the number of parameters. This also represents a 1e6-fold improvement in L_2-square error compared to standard linear FEM at comparable DoFs. Inference complexity is dominated by inner products, yielding sub-millisecond full-field predictions that scale to an arbitrary resolution. For inverse problems, KHRONOS facilitates rapid, iterative level set recovery in only a few forward evaluations, with sub-microsecond per sample latency. KHRONOS's scalability, expressivity, and interpretability open new avenues in constrained edge computing, online control, computer vision, and beyond.
Recurrent neural networks (RNNs) have been a long-standing candidate for processing of temporal sequence data, especially in memory-constrained systems that one may find in embedded edge computing environments. Recent advances in training paradigms have now inspired new generations of efficient RNNs. We introduce a streamlined and hardware-compatible architecture based on minimal gated recurrent units (GRUs), and an accompanying efficient mixed-signal hardware implementation of the model. The proposed design leverages switched-capacitor circuits not only for in-memory computation (IMC), but also for the gated state updates. The mixed-signal cores rely solely on commodity circuits consisting of metal capacitors, transmission gates, and a clocked comparator, thus greatly facilitating scaling and transfer to other technology nodes. We benchmark the performance of our architecture on time series data, introducing all constraints required for a direct mapping to the hardware system. The direct compatibility is verified in mixed-signal simulations, reproducing data recorded from the software-only network model.