Addressing the broadband gap between rural and urban regions requires rural-focused wireless research and innovation. In the meantime, rural regions provide rich, diverse use cases of advanced wireless, and they offer unique real-world settings for piloting applications that advance the frontiers of wireless systems (e.g., teleoperation of ground and aerial vehicles). To fill the broadband gap and to leverage the unique opportunities that rural regions provide for piloting advanced wireless applications, we design and implement the ARA wireless living lab for research and innovation in rural wireless systems and their applications in precision agriculture, community services, and so on. ARA focuses on the unique community, application, and economic context of rural regions, and it features the first-of-its-kind, real-world deployment of long-distance, high-capacity terrestrial wireless x-haul and access platforms as well as Low-Earth-Orbit (LEO) satellite communications platforms across a rural area of diameter over 30 km. The high-capacity x-haul platforms operate at the 11 GHz, 14 GHz, 71–86 GHz, and 194 THz bands and offer communication capacities of up to 160 Gbps at per-hop distances up to 15+ km. The wireless access platforms feature 5G-and-beyond MIMO systems operating at the 460–776 MHz, 3.4–3.6 GHz, and 27.5–28.35 GHz bands and with 14, 192, and 384 antenna elements per sector respectively, and they offer up to 3+ Gbps wireless access throughput and up to 10+ km effective cell radius. With both software-defined radios and programmable COTS systems, and through effective orchestration of these wireless resources with fiber as well as compute resources embedded end-to-end across User Equipment (UE), Base Stations (BS), edge, and cloud, including support for Bring Your Own Device (BYOD), ARA offers programmability, performance, robustness, and heterogeneity at the same time, thus enabling rural-focused co-evolution of wireless and applications while helping advance the frontiers of wireless systems in domains such as Open RAN, NextG, and agriculture applications. The resulting solutions hold the potential of reducing the rural broadband cost by a factor of 10 or more, thus making rural broadband as affordable as urban broadband. Here we present the design principles and implementation strategies of ARA, characterize its performance and heterogeneity, and highlight example wireless and application experiments uniquely enabled by ARA.

Many network faults are flooding the telecommunication companies in the form of Trouble Tickets (TT). Automation in managing these TTs is vital in increasing customer satisfaction. We develop a solution to address two challenges regarding TTs generated from fixed and mobile access network domains: prediction of resolution times and technician dispatch needs. Our study utilizes datasets from Telenor, a Swedish telecommunication operator, encompassing 35,000 access switches and 8,000 base stations. It incorporates 40,000 switch TTs and 22,000 mobile TTs during 2019-2023. None of the previous works studied multiple telecommunication domains or considered the time evolution of TTs. This work comprehensively studies several prediction models for the mentioned use cases and network domains. Our models successfully outperform the company baseline and best proposed state-of-the-art models. Within 1-hour confidence interval, our method can correctly predict shortest ranges of resolution times for 90% of switch TTs and 80% of mobile TTs. We also predict the necessity of dispatching workforce to the place with weighted F1 scores of respectively, 88% and 89% for switch and mobile TTs which shows high average accuracy of our system in prediction across both dispatch and non-dispatch TT classes to assist operation. With these scores, our model is capable of allocating resources automatically, enhancing customer satisfaction. We also studied the TTs evolution, for example, for switch TTs, within 15 minutes of creation time, prediction improves by 57% and 50%, for resolution and dispatch prediction, respectively.

With the increasing focus on flexible automation, which emphasizes systems capable of adapting to varied tasks and conditions, exploring future deployments of cloud and edge-based network infrastructures in robotic systems becomes crucial. This work, examines how wireless solutions could support the shift from rigid, wired setups toward more adaptive, flexible automation in industrial environments. We provide a quality of control (QoC) based abstraction for robotic workloads, parameterized on loop latency and reliability, and jointly optimize system performance. The setup involves collaborative robots working on distributed tasks, underscoring how wireless communication can enable more dynamic coordination in flexible automation systems. We use our abstraction to optimally maximize the QoC ensuring efficient operation even under varying network conditions. Additionally, our solution allocates the communication resources in time slots, optimizing the balance between communication and control costs. Our simulation results highlight that minimizing the delay in the system may not always ensure the best QoC but can lead to substantial gains in QoC if delays are sometimes relaxed, allowing more packets to be delivered reliably.

Low-power wireless Internet of Things (IoT) devices employ Time Slotted Channel Hopping (TSCH) Medium Access Control to achieve predictable timing behaviour. TSCH aims at collision-free scheduling by exploiting diversity over time (slots) and frequency (channels). However, existing works on performance and worst-case analysis are based on deterministic models, which lead to rather pessimistic non-realistic results, i.e. tools for probabilistic performance analysis of TSCH schedulers are still lacking. In this context, we devised a Stochastic Network Calculus model that enables to calculate end-to-end delays for specific traffic flows and (deadline) violation probability, building on Moment Generating Functions. We instantiate this SNC model and provide bounds for three widely used TSCH schedulers, namely Minimal Scheduling Function, Orchestra, and a custom collision-free scheduler, with different parameters such as radio duty-cycle, radio link quality, and traffic arrival rate. We demonstrate that our proposed model closely follows the simulation results, under different network scenarios.

Supporting applications in emerging domains like cyber-physical systems and human-in-the-loop scenarios typically requires adherence to strict end-to-end delay guarantees. Contributions of many tandem processes unfolding layer by layer within the wireless network result in violations of delay constraints, thereby severely degrading application performance. Meeting the application's stringent requirements necessitates coordinated optimization of the end-to-end delay by fine-tuning all contributing processes. To achieve this task, we designed and implemented EDAF, a framework to decompose packets' end-to-end delays and determine each component's significance for 5G network. We showcase EDAF on OpenAirInterface 5G uplink, modified to report timestamps across the data plane. By applying the obtained insights, we optimized end-to-end uplink delay by eliminating segmentation and frame-alignment delays, decreasing average delay from 12ms to 4ms.

In this paper, we consider an Internet of Things (IoT) network supporting latency-critical transmissions to multiple nodes in the finite blocklength regime. An energy efficiency maximizing design is provided via efficiently jointly allocating the power and blocklength among the transmissions to different nodes, while guaranteeing per-node constraint of reliability. To address the formulated non-convex problem, we propose a Majorization-Minimization-based approach. Specifically, we tightly approximate the problem at each local point by introducing an auxiliary constant. Then, we rigorously prove the convexity of the conducted local problem. By successively optimally solving each convex local problem, a near-optimal solution is finally achieved. Via simulations, we demonstrate the performance advantages of the proposed joint design.

We consider a resource-constrained Edge Device (ED), such as an IoT sensor or a microcontroller unit, embedded with a small-size ML model (S-ML) for a generic classification application and an Edge Server (ES) that hosts a large-size ML model (L-ML). Since the inference accuracy of S-ML is lower than that of the L-ML, offloading all the data samples to the ES results in high inference accuracy, but it defeats the purpose of embedding S-ML on the ED and deprives the benefits of reduced latency, bandwidth savings, and energy efficiency of doing local inference. In order to get the best out of both worlds, i.e., the benefits of doing inference on the ED and the benefits of doing inference on ES, we explore the idea of Hierarchical Inference (HI), wherein S-ML inference is only accepted when it is correct, otherwise the data sample is offloaded for L-ML inference. However, the ideal implementation of HI is infeasible as the correctness of the S-ML inference is not known to the ED. We thus propose an online meta-learning framework that the ED can use to predict the correctness of the S-ML inference. In particular, we propose to use the probability corresponding to the maximum probability class output by S-ML for a data sample and decide whether to offload it or not. The resulting online learning problem turns out to be a Prediction with Expert Advice (PEA) problem with continuous expert space. For a full feedback scenario, where the ED receives feedback on the correctness of the S-ML once it accepts the inference, we propose the HIL-F algorithm and prove a sublinear regret bound √ n ln(1/λ min )/2 without any assumption on the smoothness of the loss function, where n is the number of data samples and λ min is the minimum difference between any two distinct maximum probability values across the data samples. For a no-local feedback scenario, where the ED does not receive the ground truth for the classification, we propose the HIL-N algorithm and prove that it has O ( n 2/3 ln 1/3 (1/λ min )) regret bound. We evaluate and benchmark the performance of the proposed algorithms for image classification application using four datasets, namely, Imagenette and Imagewoof [1], MNIST [2], and CIFAR-10 [3].

Deep learning (DL) applications have rapidly evolved to address increasingly complex tasks by leveraging large-scale, resource-intensive models. However, deploying such models on low-power devices is not practical or economically scalable. While cloud-centric solutions satisfy these computational demands, they present challenges in terms of communication costs and latencies for real-Time applications when every computation task is offloaded. To mitigate these concerns, hierarchical inference (HI) frameworks have been proposed, enabling edge devices equipped with small ML models to collaborate with edge servers by selectively offloading complex tasks. Existing HI approaches depend on immediate offloading of data upon selection, which can lead to inefficiencies due to frequent communication, especially in time-varying wireless environments. In this work, we introduce Batch HI, an approach that offloads samples in batches, thereby reducing communication overhead and improving system efficiency while achieving similar performance as existing HI methods. Additionally, we find the optimal batch size that attains a crucial balance between responsiveness and system time, tailored to specific user requirements. Numerical results confirm the effectiveness of our approach, highlighting the scenarios where batching is particularly beneficial.

Nature-based climate solutions supply carbon credits generated from net carbon drawdown in exchange for project funding, but their credibility is challenged by the inherent variability and impermanence of drawdown. By evaluating drawdown benefits from a social cost of carbon perspective, project developers can enhance credibility and estimate impermanence by conservatively anticipating drawdowns to be eventually released following a release schedule, issuing additional credits when actual release is less severe than anticipated. We demonstrate how we can use ex post observations of drawdowns to construct optimal release schedules that limit the risk of credit reversals (when net drawdown is negative). We simulate both theoretical and real-life projects to examine how this approach balances the trade-off between generating credits evaluated as more permanent and limiting reversal risk. We discuss how this approach incentivizes project performance and provides a pragmatic solution to challenges facing larger-scale implementation of nature-based climate solutions.