Papers for

edge network operators

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Client selection improves federated multimodal model fine-tuning under network limits

CARE-FL: Federated Fine-Tuning of Multimodal LLMs in Resource-Constrained Edge Networks

Abstract: Domain-specific adaptation of Multimodal Large Language Models (MLLMs) in privacy-sensitive edge environments requires federated fine-tuning, yet client selection remains challenging due to heterogeneous multimodal data and modality availability, dynamic system conditions, and evolving client utility during training. We propose Constraint-Aware Reinforcement Learning for Efficient Federated Fine-Tuning (CARE-FL), a constrained online client-selection framework for federated MLLM Low-Rank Adaptation (LoRA) fine-tuning. CARE-FL formulates client participation as a sequential decision process and employs Q-value decomposition to address the large combinatorial joint action space through scalable client-level long-term utility estimation. These utilities are learned online using a modified Double Deep Q-Learning (DDQL) scheme, while constraint-aware $ε$-greedy execution selects feasible client subsets under per-round latency and bandwidth constraints. Complementing client selection, an adaptive aggregation mechanism further controls potentially unstable global updates and places greater weight on the previous model when needed. Experiments on the CrisisMMD disaster-response task in emulated heterogeneous wireless edge environments show that CARE-FL achieves higher and more stable test accuracy than the compared client-selection baselines across different system settings, while satisfying the imposed latency and bandwidth constraints.

Wed 4 NovDistributed, Parallel, and Cluster Computing
The gist
Training large language AI models that understand images and text together is hard when data is spread across different devices that have different connection speeds and data types. The paper offers a new way to pick which devices should help train the model so the process is faster and uses less network data. Their method learns over time which devices to choose, balancing speed and bandwidth limits. Tests show it gives better and more stable AI performance on disaster response tasks than older methods.
Open → 2611.06169v1

Replica aware recovery lowers latency in split inference on IoT devices

Replica-Aware Split Inference for Edge IoT Systems

Abstract: Split Inference (SI) addresses the hardware resource limitations of Internet of Things (IoT) devices by distributing tiny Machine Learning (TinyML) model segments across multiple IoT devices. However, a device failure can interrupt the inference pipeline. This paper proposes a replica-aware recovery approach that deploys replicas of TinyML model segments originally hosted on failed devices onto surviving devices. We formulate the joint selection of replica hosts and communication protocols as a mixed-integer linear programming (MILP) problem that minimizes post-reconfiguration inference latency, accounting for computation and transmission latencies, under flash memory and runtime memory constraints. Simulations using MobileNetV2 on heterogeneous IoT networks compare the proposed approach with four baselines. We implement the model in a network of 15 IoT devices with a maximum link-rate reduction of 75%, and show that MILP approach achieves an average modeled end-to-end inference latency that is 6.8% lower than that of Computation Greedy and 18.3% lower than that of Random. Index Terms TinyML, split inference, Internet of Things, fault tolerance, replica placement.

Tue 3 NovDistributed, Parallel, and Cluster ComputingArtificial IntelligenceSoftware Engineering
The gist
Devices in the Internet of Things are small and can't run big AI models alone, so they split the task across many devices. When one device fails, the process stops working. The paper shows a way to recover by copying the failed device’s model parts onto others, choosing hosts and communication methods to keep the whole process fast. They tested this using a real AI model and found their approach reduces delays better than other common methods.
Open → 2611.04779v1

Edge task placement reduces server overload risk under uncertain workloads

Minimizing Bottleneck Overload Risk in Online Edge Task Placement under Workload Uncertainty

Abstract: In edge computing, tasks must be placed on servers of finite capacity before workloads are realized, often with only workload means and variances known. An assignment balanced on expected load can still leave a server with more fluctuations than its residual capacity can absorb. Per-server risk constraints can enforce a prescribed overload tolerance but do not distinguish the assignments that meet it by how close the riskiest server comes to the limit. We show that, for independent tasks, placement preserves a period's total variance and total residual capacity, and decides only how fluctuation, measured by standard deviation, is paired with residual capacity across servers. The risk ratio, the ratio of fluctuation to residual capacity, parameterizes a server's worst-case overload probability over all aggregate workload distributions with given mean and variance. We minimize the largest risk ratio, attained at the risk bottleneck. Risk-Bottleneck Placement (RBP) places tasks in descending order of standard deviation, each on the server with the smallest post-assignment risk ratio, and refines the assignment with relocations and swaps within a finite per-period online budget. On small instances solved exactly by enumerating every assignment, minimizing the maximum expected utilization leaves the largest risk ratio 25.66-29.19% above its exact minimum on average, even with risk-minimizing tie-breaking, and the corresponding gap for RBP is 1.77-2.64%. In trace-derived multi-period simulations at two workload pressures, RBP reduces the worst-server overload probability by 26.0% and 18.0% relative to the best baseline and remains the lowest at every tested pressure and scale.

Tue 3 NovDistributed, Parallel, and Cluster Computing
The gist
In edge computing, servers have limited capacity, and tasks must be assigned before knowing the exact workload. The authors show that just balancing expected workloads isn’t enough because fluctuations can cause overloads. They propose a method called Risk-Bottleneck Placement (RBP) that assigns tasks to minimize the worst-case overload risk across servers. Tests show RBP cuts the highest overload chance by around 18-26% compared to other methods, improving system reliability.
Open → 2611.04425v1

Edge server manages IoT memory to save energy and keep data fresh

Energy-Efficient Timely Relevant IoT Data Retrieval with Wireless Memory Approximation

Abstract: Collecting relevant data from Internet of Things devices under freshness and energy constraints is a key challenge in 6G edge networks. A lightweight neural network on the device filters irrelevant observations, but storing it in dynamic random access memory consumes refresh energy during standby. This letter proposes a pull-based framework where an edge server remotely activates and approximates memory and jointly selects the activation timing, refresh period, model size, and transmission threshold. We derive the expected freshness and energy over erasure channels in closed form and show that the scheme consumes less energy than an always-on baseline under the same freshness constraint.

Wed 21 OctNetworking and Internet Architecture
The gist
Collecting timely and useful data from Internet of Things (IoT) devices while saving energy is tough because devices need power to keep important data ready. The authors propose a way where the edge server decides when to wake up devices and approximates stored memory, reducing energy use without losing data freshness. Their method outperforms always-on devices by balancing when and how data is refreshed and transmitted. This makes IoT data collection more energy-efficient.
Open → 2610.29236v1

Stacked intelligent metasurfaces improve edge computing communication efficiency

Stacked Intelligent Metasurface-Based MultiModal Feature Fusion for Collaborative Inference in Edge Networks

Abstract: In navigating the challenges of computational and communication overhead within the 6G-era Collaborative Inference (CI), this paper explores the feasibility of employing Stacked Intelligent Metasurface (SIM) to perform CI, termed SIM-CI. The SIM functions as a physical computing unit, receiving multi-user Over-The-Air (OTA) aggregated features within the same spectrum. It thus executes the necessary spatial-frequency combining and inference in a single physical wave propagation.SIM-CI also incorporates a tri-heterogeneity robustness adapter to address the network heterogeneity, such as multi-modal data and user inference models, as well as a quality-aware contrastive approach that align features from different modalities and extraction levels into a unified feature space. Considering the heterogeneous and dynamic channels, we further integrate Channel State Information (CSI) as an additional input into the adapter design, enabling users' early perception of potential channel states. Moreover, to maximally mitigate the impact of channel variations, we design a new hybrid beamforming network that combines the robustness of model-driven pre-equalization with the adaptability of data-driven learning. Experimental results demonstrate the effectiveness of our SIM-CI compared to digital-domain inference, achieving 98.6\% accuracy to conversional digital-domain CI approaches under various heterogeneous conditions, while reducing computational energy by 25\% and requiring 75\% fewer transmission time slots.

Fri 9 OctDistributed, Parallel, and Cluster Computing
The gist
Communicating and computing data from multiple users at the same time can be slow and energy-consuming. The authors propose using a special layered surface, called a stacked intelligent metasurface, that processes combined signals physically through waves, saving time and energy. Their design also adapts to different data types, user devices, and changing network conditions to keep accuracy high. Tests show this approach works nearly as well as traditional digital methods but uses less power and fewer communication steps.
Open → 2610.13364v1

Adaptive client clustering improves federated learning in edge networks

Adaptive Client Clustering and Coordination for Federated Learning Workflow Management in Edge Networks

Abstract: Federated learning (FL) is increasingly deployed as a managed learning service rather than as a set of isolated training jobs. In networked edge environments, dependent FL service flows must coordinate heterogeneous clients, non-IID data, fluctuating communication latency, and precedence-constrained tasks under service-level completion requirements. These coupled factors make participant management central to both time-totarget performance and learning stability. This paper proposes A-CoDa, an adaptive clustered coordination framework for managing dependent FL flows. A-CoDa first uses label-distribution divergence (LDD)-based greedy-balanced clustering to construct statistically coherent and size-aware client groups, which serve as a scalable management abstraction. Building on this structure, we design FedMIX, an uncertainty-aware intra-/inter-cluster participation mechanism that ranks clients by a loss-latency-uncertainty utility and adaptively controls cross-cluster probing according to training progress and latency conditions. A dependency-aware DAG scheduler then orchestrates layer-wise task execution so that parallelism and precedence constraints are jointly respected. We further provide a convergence analysis that frames the result as a sufficient loss-domain design bound, explicitly relating the attainable error floor and sufficient communication rounds to LDDinduced sampling mismatch, residual distribution shift, local-SGD drift, stochastic variance, and adaptive probing budgets. Experiments on handwriting, wearable-sensing, product-image, and medical-imaging tasks evaluate A-CoDa under dependent FL workflows and demonstrate its effectiveness in reducing end-toend completion time while maintaining competitive accuracy.

Sun 27 SeptDistributed, Parallel, and Cluster Computing
The gist
Federated learning lets many devices work together to train a machine learning model without sharing their data directly, but this gets tricky when devices differ a lot and network delays vary. The authors propose A-CoDa, a method that groups similar devices into clusters and carefully decides which ones should participate in learning steps to speed up the process and keep the model accurate. They also design a scheduler to manage tasks efficiently, respecting the order needed for learning to progress. Tests on various real-world problems show that their approach cuts down training time while maintaining good results.
Open → 2609.33544v1

Parallel optimization method speeds up learning in edge networks

P-GADMM: Parallel Group-Based ADMM for Asynchronous Optimization in Heterogeneous Edge Networks

Abstract: The Alternating Direction Method of Multipliers (ADMM) is widely used for distributed optimization, but its synchronous implementation can suffer from efficiency loss in heterogeneous edge networks, where fast clients or groups need to wait for slower ones before global updates can be completed. Existing group-based ADMM methods reduce communication overhead through grouping, but their grouping rules usually focus on data similarity or network topology and do not explicitly account for computation heterogeneity. To address this issue, this paper proposes Parallel Group-Based ADMM (P-GADMM) for distributed optimization in heterogeneous edge networks. P-GADMM forms computation-aware edge groups according to client computational capabilities and local data sizes, which reduces training-speed variation within each group. It further combines edge-level aggregation with bounded asynchronous coordination at the cloud, allowing active groups to participate in global updates without waiting for slower groups while controlling stale group information through a delay threshold. For strongly convex group objectives, we establish convergence guarantees for an idealized form of P-GADMM under bounded group-level staleness, showing a time-averaged convergence behavior up to a staleness-induced asymptotic error neighborhood. Experiments show that P-GADMM reduces wall-clock training time compared with representative baselines while maintaining comparable final accuracy.

Thu 17 SeptDistributed, Parallel, and Cluster Computing
The gist
In networks where many devices work together to learn from data, slower devices can hold up the whole process. The authors created a new method called P-GADMM that groups devices based on their speed and data size, so faster groups can update more quickly without waiting for the slowest ones. This method uses a smart way to combine results with some delay allowed, keeping the learning accurate while saving time. Tests showed it finishes training faster than other methods with similar results.
Open → 2609.20006v1

Edge network security improved by governor checks on automated actions

Autonomy in Check: Governor-Mediated Adaptive Security at the Edge

Abstract: Adaptive security at the network edge increasingly relies on automated planners, including rule-based controllers, learned policies, and LLM-assisted agents, that translate observations into enforcement actions. Once such a planner can influence live policy state, syntactic validity is not enough. A semantically wrong action, produced from incomplete or manipulated observations, can be faithfully executed by an enforcement substrate that cannot judge mission context. We address this problem by treating the boundary between planner output and kernel enforcement input as the primary security object. We propose a split-control architecture in which an untrusted planner emits typed security intents, a deterministic governor checks each intent against safety, resource, temporal-stability, and proportionality invariants, and only admitted actions are bound to signed receipts and compiled into pre-installed eBPF map updates. The paper formalizes this trust-boundary problem, defines three threat classes, develops the governor admission predicate, and reports an end-to-end prototype. Across rule-based and LLM-assisted planners on a Raspberry Pi 5 testbed connected to the university 5G Test Network, the governor admits, rejects, and bounds intents at microsecond cost without disrupting protected-flow regularity. The contribution is conceptual as much as empirical: adaptive security does not need to trust the author of an action. It needs a mediation boundary that decides whether the action is admissible.

Wed 16 SeptCryptography and SecurityArtificial Intelligence
The gist
Security systems at the edge of networks use automatic planners to decide actions, but these planners can make mistakes or be tricked by bad information. The paper shows that just checking if an action looks correct on the surface is not enough. Instead, the authors propose a middleman called a governor that carefully checks each action before it is carried out to make sure it follows safety and fairness rules. They tested this approach on small devices connected to a 5G network and found it works very quickly without messing up normal network flows.
Open → 2609.18338v1