Artificial Intelligence (AI) is being incorporated in several optimization, scheduling, orchestration as well as in native communication network functions. This paradigm shift results in increased energy consumption, however, quantifying the end-to-end energy consumption of adding intelligence to communication systems remains an open challenge since conventional energy consumption metrics focus on either communication, computation infrastructure, or model development. To address this, we propose a new metric, the Energy Cost of AI Lifecycle (eCAL) of an AI model in a system. eCAL captures the energy consumption throughout the development, deployment and utilization of an AI-model providing intelligence in a communication network by (i) analyzing the complexity of data collection and manipulation in individual components and (ii) deriving overall and per-bit energy consumption. We show that as a trained AI model is used more frequently for inference, its energy cost per inference decreases, since the fixed training energy is amortized over a growing number of inferences. For a simple case study we show that eCAL for 100 inferences is 2.73 times higher than for 1000 inferences. Additionally, we have developed a modular and extendable opensource simulation tool to enable researchers, practitioners, and engineers to calculate the end-to-end energy cost with various configurations and across various systems, ensuring adaptability to diverse use cases.

Modern edge-cloud systems face challenges in ef-ficiently scaling resources to handle dynamic and unpredictable workloads. Traditional scaling approaches typically rely on static thresholds and predefined rules, which are often inadequate for optimizing resource utilization and maintaining performance in distributed and dynamic environments. This inefficiency hin-ders the adaptability and performance required in edge-cloud infrastructures, which can only be achieved through the newly proposed in-place scaling. To address this problem, we propose the Multi-Agent Reinforcement Learning-based In-place Scaling Engine (MARLISE) that enables seamless, dynamic, reactive control with in-place resource scaling. We develop our solution using two Deep Reinforcement Learning algorithms: Deep Q-Network (DQN), and Proximal Policy Optimization (PPO). We analyze each version of the proposed MARLISE solution using dynamic workloads, demonstrating their ability to ensure low response times of microservices and scalability. Our results show that MARLISE- based approaches outperform heuristic method in managing resource elasticity while maintaining microservice response times and achieving higher resource efficiency.

In this paper, we demonstrate a smart resource auto-scaling solution based on Multi-Agent Deep Reinforcement Learning (MADRL). Our decentralized approach dynamically adapts compute resources to services to ensure a smooth response to fluctuating user demands. The demonstration, consisting of a visualization and control dashboard, a cloud edge hardware configuration and an AI-based wireless fingerprinting service, shows in-place real-time scaling of resources in the virtual environment and its impact on service response time and performance.

Time‐series classification is a relevant step supporting decision‐making processes in various domains, and deep neural models have shown promising performance in this respect. Despite significant advancements in deep learning, the theoretical understanding of how and why complex architectures function remains limited, prompting the need for more interpretable models. Recently, the Kolmogorov–Arnold Networks (KANs) have been proposed as a more interpretable alternative to deep learning. While KAN‐related research is significantly rising, to date, the study of KAN architectures for time‐series classification has been limited. In this paper, we aim to conduct a comprehensive and robust exploration of the KAN architecture for time‐series classification utilizing 117 datasets from UCR benchmark archive, from multiple different domains. More specifically, we investigate (a) the transferability of reference architectures designed for regression to classification tasks, (b) the hyperparameter and implementation configurations for an architecture that best generalizes across 117 datasets, (c) the associated complexity trade‐offs, and (d) KANs interpretability. Our results demonstrate that (1) the Efficient KAN outperforms MLPs in both performance and training times, showcasing its suitability for classification tasks. (2) Efficient KAN exhibits greater stability than the original KAN across grid sizes, depths, and layer configurations, especially when lower learning rates are employed. (3) KAN achieves competitive accuracy compared to state‐of‐the‐art models such as HIVE‐COTE2 and InceptionTime, while maintaining smaller architectures and faster training times, highlighting its favorable balance of performance and transparency. (4) The interpretability of the KAN model, as confirmed by SHAP analysis, reinforces its capacity for transparent decision‐making.

With the process of democratization of the network edge, hardware and software for networks are becoming available to the public, overcoming the confines of traditional cloud providers and network operators. This trend, coupled with the increasing importance of AI in 6G and beyond cellular networks, presents opportunities for innovative AI applications and systems at the network edge. While AI models and services are well-managed in cloud systems, achieving similar maturity for serving network needs remains an open challenge. Existing open solutions are emerging and are yet to consider democratization requirements. In this work, we identify key requirements for democratization and propose NAOMI, a solution for democratizing AI/ML workflows at the network edge designed based on those requirements. Guided by the functionality and overlap analysis of the O-RAN AI/ML workflow architecture and MLOps systems, coupled with the survey of open-source AI/ML tools, we develop a modular, scalable, and distributed hardware architecture-independent solution. NAOMI leverages state-of-the-art open-source tools and can be deployed on distributed clusters of heterogeneous devices. The results show that NAOMI performs up to 40% better in deployment time and up to 73% faster in AI/ML workflow execution for larger datasets compared to AI/ML Framework, a representative open network access solution, while performing inference and utilizing resources on par with its counterpart.

The shift towards software-centric network infrastructures is driven by the increasing need for networks to be responsive, flexible, and scalable in the face of an expanding set of connected devices. The digital twin (DT) approach, mirroring physical entities in a digital format, has emerged as a key enabler of network reliability and availability. Incorporating artificial intelligence (AI) into DTs enhances the resilience of networks by providing in-depth analysis and increasingly automated mitigation strategies against operational disruptions. In this work, we propose a new AI-based information extraction module that is part of the DT Monitoring component able to process RSS data, extract and characterize abrupt anomalies. The output of this component is used to maintain an anomaly history in the Link Abstraction within the DT and subsequently inform possible automatic mitigation actions. We design the AI-based information extraction module to identify and characterize three types of RSS based anomalies. Our extensive performance analysis on 10 versions of the “You Only Look Once” architecture reveals that YOLOv8n produces a good tradeoff between performance and computational complexity. We show that our approach performs on par with the state of the art for anomaly detection, while enabling anomaly characterization by location, duration, and severity. By employing two SotA explainability algorithms, we also provide insights into the important regions of the input that trigger the selected model’s classification and characterization decisions.

Energy management systems (EMS), as enablers of more efficient energy consumption, monitor and manage appliances to help residents be more energy efficient and thus more frugal. Recent appliance detection and identification techniques for such systems rely on machine learning. However, machine learning solutions for appliance classification on existing low-frequency household metering have not yet been thoroughly investigated. In this paper, we propose CARMEL, a new approach for identifying home appliances from load monitoring in building EMS based on a new data representation technique and a new model that leverages spatio-temporal correlations in the new representation. The proposed data representation technique performs dimensionality expansion of time series that scales linearly, rather than quadratically and, together with the proposed model, outperform the state of the art image transformation models by 5 percentage points. Evaluation on 5 different low-frequency household metering datasets, considering 29 appliances in total, shows that the proposed representation and the corresponding resource-aware deep learning architecture (1) achieve an average weighted F1 score of 0.92 and (2) require only 230 labeled samples and 3x fewer epochs to transfer to new households.

The rise of cloud computing has been enabled by advances in virtualization and containerization technology. Over the past decade, the use of cloud computing has grown rapidly and has had a significant impact on digital transformation with many enterprises migrating to public clouds. While convenient and cost efficient, such approaches are prone to certain data privacy, compliance and security risks. The ongoing democratisation of cloud technologies represented by the increasing number of open source projects, has enabled certain enterprises to easily develop their on-premise cloud infrastructure. However, these open source projects are largely enterprise level and still too complex for small and micro enterprises and academic environments. To further decrease the on-premise infrastructure deployment and management barrier, we first provide an analysis of the existing on-premises PaaS workflows and solutions, along with the complexity of their deployment models, and identify the requirements for simple PaaS solutions for small environments with limited resources. We then introduce Kubitect as an enabler of on-premises PaaS democratization and expedite digital transformation. Kubitect is a lightweight single file declarative infrastructure configuration solution for on-premises cluster definition, instantiation and update. Our qualitative and quantitative evaluation shows the advantage of Kubitect for small environments where simplicity is more important than deploy time, assuming the latter is relatively comparable with alternative solutions.

Nowadays, modern man-made infrastructures are being upgraded with information and communication technologies that form large wireless networks. Such large wireless networks must be monitored to ensure reliable operation by the proactive detection and correction of link failures or abnormal network behaviour in view of uninterrupted business operations. In this paper, we present a new method for detecting wireless link anomalies based on graph neural networks. The proposed method transforms time series data into graphs using a Markov Transition Field transformation. The data resulting from the transformation then trains a new graph neural network architecture to learn to successfully discriminate between 4 different link layer anomalies with an average F1 score of 0.956. The resulting model achieves competitive results with superior detection ability for the more subtle slow degradation anomaly while having up to ≈230 times fewer trainable parameters compared to the state-of-the-art, which subsequently makes it computationally much more efficient.

Driven by various academic, standardization and regulatory initiatives, recent research on spectrum resource utilisation has focused also on technology and transmission classification using various deep learning (DL) architectures. However, especially in unlicensed bands it is often hard to obtain labelled data of sufficient quality for training DL for all transmissions that may occur. Therefore in this paper we investigate clustering techniques that require no labelled data or prior knowledge on continuous spectrum sensing sweeps over a 200 kHz wide band in the unlicensed European 868 MHz frequency band. Using several clusterability tests we show that the sweeps can be clustered, however the number of clusters is not clear. By analyzing the sweeps with three state-of-the-art techniques, K-Means, Agglomerative Hierarchical Clustering, and Hierarchical Density-Based Spatial Clustering (HDBSCAN) 9-10 clusters are discovered. The quantitative evaluation shows that HDBSCAN outperforms the other two and the qualitative analysis shows that HDBSCAN seems to be able to better discriminate between the same technology transmitting at/from different power levels/distance, however it has a relatively poorer class balance compared to K-Means and tends to group more transmissions than needed in the majority no transmission cluster.

Time series classification is an important task in many fields. In intrusive and non-intrusive load monitoring (N)ILM, time series data are obtained from power measurements of electrical appliances that are not known in advance, therefore extracting the type of appliance from the data is a relevant problem in smart grids. We propose a transformation that encodes a time series into an image that can be effectively recognised by well known image classification algorithms. The transformation is based on plotting the time series into a matrix and filling the area under it, producing a more pronounced representation of its shape. We perform an extensive evaluation on 1 synthetic and 4 measured datasets. Our experiments on synthetic and mixed measured data yielded F1 scores of 99.2% and 85.9%, respectively, and outperformed the state-of-the-art on three out of four tested datasets. Additionally, we conclude that our method tends to work better on longer time series segments, as the resulting images contain more distinguishing features.

In recent years, much work has been done on processing of wireless spectrum data involving machine learning techniques in domain-related problems for cognitive radio networks, such as anomaly detection, modulation classification, technology classification and device fingerprinting. Most of the solutions are based on labeled data, created in a controlled manner and processed with supervised learning approaches. However, spectrum data measured in real-world environment is highly nondeterministic, making its labeling a laborious and expensive process, requiring domain expertise, thus being one of the main drawbacks of using supervised learning approaches in this domain. In this paper, we investigate the utilization of self-supervised learning (SSL) for exploring spectrum activities in a real-world unlabeled data. In particular, we assess the performance of SSL models, based on the reference DeepCluster architecture. We carefully consider the current state-of-the-art feature extractors, taking into account the performance and complexity trade-offs. Our findings demonstrate that SSL models achieve superior performance regarding the feature quality and clustering performance compared to baseline feature learning approaches. With SSL models we achieve significant reduction of the feature vectors size by two orders of magnitude, while improving the performance by a factor ranging from 2 to 2.5 across the evaluation metrics, supported by visual assessment. Furthermore, we showcase how adapting the reference SSL architecture to domain-specific data is followed by a substantial reduction in model complexity up to one order of magnitude, without compromising, and in some cases, even improving the clustering performance.

Today, there are billions of connected IoT devices and their number continues to grow as they contribute to the digitalization of infrastructures. However, the deployment process of these smart wireless devices when delivered to customer premises is slow and error prone as each of them needs to be provisioned with authentication credentials to access the corporate network. In this paper, we propose HANNA, a human-friendly provisioning and configuration framework for smart devices, that extends the zero-touch paradigm to large IoT deployments by introducing voice assisted configuration in combination with large scale ad-hoc communications to overcome the initial installation effort of IoT deployments. The most prominent role in HANNA is played by the assisting device, which includes a voice assistant capable of correctly understanding a minimum number of keywords required for initial provisioning and configuration of the devices. The device’s role is to interact with the user and ensure that all provisioning details are received. These are then converted into appropriate machine instructions for further use by the mass provisioning mechanism. We provide an example prototype implementation of HANNA and evaluate the performance of the assisting device in the human-to-machine communication phase and the performance of the selected communication technique in the machine-to-machine communication phase. Our results show the potential of existing speech-to-text engines for this application area and also reveal shortcomings with respect to the robustness of the engines in office-like working environments as well as with respect to user’s gender and language proficiency level. Additionally we show that the proposed machine-to-machine provisioning approach is always faster compared to manual provisioning for cases with more than ten devices.

The number of end devices that use the last-mile wireless connectivity is dramatically increasing with the rise of smart infrastructures and requires reliable functioning to support smooth and efficient business processes. To efficiently manage such massive wireless networks, more advanced and accurate network monitoring and malfunction detection solutions are required. In this article, we perform a first-time analysis of image-based representation techniques for wireless anomaly detection using recurrence plots (RPs) and Gramian angular fields and propose a new deep learning architecture enabling accurate anomaly detection. We elaborate on the design considerations for developing a resource-aware architecture and propose a new model using time series to image transformation using RPs. We show that the proposed model: 1) outperforms the one based on Gramian angular fields by up to 14% points; 2) outperforms classical ML models using dynamic time warping by up to 24% points; 3) outperforms or performs on par with mainstream architectures, such as AlexNet and VGG11 while having their weights and up to ≈8% of their computational complexity; and d) outperforms the state of the art in the respective application area by up to 55% points. Finally, we also explain on randomly chosen examples how the classifier takes decisions.

Location based services, already popular with end users, are now inevitably becoming part of new wireless infrastructures and emerging business processes. The increasingly popular Deep Learning (DL) artificial intelligence methods perform very well in wireless fingerprinting localization based on extensive indoor radio measurement data. However, with the increasing complexity these methods become computationally very intensive and energy hungry, both for their training and subsequent operation. Considering only mobile users, estimated to exceed 7.4 billion by the end of 2025, and assuming that the networks serving these users will need to perform only one localization per user per hour on average, the machine learning models used for the calculation would need to perform 65×1012predictions per year. Add to this equation tens of billions of other connected devices and applications that rely heavily on more frequent location updates, and it becomes apparent that localization will contribute significantly to carbon emissions unless more energy-efficient models are developed and used. This motivated our work on a new DL-based architecture for indoor localization that is more energy efficient compared to related state-of-the-art approaches while showing only marginal performance degradation. A detailed performance evaluation shows that the proposed model produces only 58% of the carbon footprint while maintaining 98.7% of the overall performance compared to state of the art model external to our group. Additionally, we elaborate on a methodology to calculate the complexity of the DL model and thus the CO2footprint during its training and operation.

Machine learning (ML) techniques play a significant role in detecting anomalous wireless links. However, to date, to the extent of our knowledge, there is no robust classifier that would work in a realistic scenario where various anomalies could appear concurrently in the time-series gleaned from the network monitoring tools. In this paper, we propose a new deep learning based classifier and show that is able to outperform the state of the art for existing link layer anomalies. Our evaluation results demonstrate that the state-of-the ML models perform with an average accuracy of about 63%, whereas the average accuracy of the proposed DL model is around 90%, indicating a significant improvement of 27 percentage points anomaly detection performance.

The Internet of Things (IoT) is being widely adopted in today's society, interconnecting smart embedded devices that are being deployed for indoor and outdoor environments, such as homes, factories and hospitals. Along with the growth in the development and implementation of these IoT devices, their simple and rapid deployment, initial configuration and out-of-the-box operational provisioning are becoming prominent challenges to be circumvented. Considering a large number of heterogeneous devices to be deployed within next generation IoT networks, the amount of time needed for manual provisioning of these IoT devices can significantly delay the deployment and manual provisioning may introduce human-induced failures and errors. By incorporating zero-touch provisioning (ZTP), multiple heterogeneous devices can be provisioned with less effort and without human intervention. In this paper, we propose software-enabled access point (Soft-AP)- and Bluetooth-based ZTP solutions relying only on a single mediator device and evaluate their performances using LOG-A-TEC testbed against manual provisioning in terms of the time required for provisioning (time-to-provision, TTP). We demonstrate that on average, Soft-AP- and Bluetooth-based ZTP solutions outperform manual provisioning with about 154% and 313% when compared to the expert provisioning, and with about 434% and 880% when compared to the non-expert provisioning in terms of TTP performances, respectively.

Ensuring a reliable communication in wireless networks strictly depends on the effective estimation of the link quality, which is particularly challenging when propagation environment for radio signals significantly varies. In such environments, intelligent algorithms that can provide robust, resilient and adaptive links are being investigated to complement traditional algorithms in maintaining a reliable communication. In this respect, the data-driven link quality estimation (LQE) using machine learning (ML) algorithms is one of the most promising approaches. In this paper, we provide a quantitative evaluation of design decisions taken at each step involved in developing a ML-based wireless LQE on a selected, publicly available dataset. Our study shows that, re-sampling to achieve training class balance and feature engineering have a larger impact on the final performance of the LQE than the selection of the ML method on the selected data.

In this paper we propose an architecture for fully-reconfigurable, plug-and-play wireless sensor network testbed. The proposed architecture is able to reconfigure and support easy experimentation and testing of standard protocol stacks (i.e. uIPv4 and uIPv6) as well as non-standardized clean-slate protocol stacks (e.g. configured using RIME). The parameters of the protocol stacks can be remotely reconfigured through an easy to use RESTful API. Additionally, we are able to fully reconfigure clean-slate protocol stacks at run-time. The architecture enables easy set-up of the network - plug - by using a protocol that automatically sets up a multi-hop network (i.e. RPL protocol) and it enables reconfiguration and experimentation - play - by using a simple, RESTful interaction with each node individually. The reference implementation of the architecture uses a dual-stack Contiki OS with the ProtoStack tool for dynamic composition of services.

Spectrum sensing is one of the core functionalities of a true cognitive radio (CR) that supports operation over a broad range of frequencies and can autonomously adapt transmission parameters to the operating environment. There are several types of hardware ranging from sophisticated (i.e. Nutaq Radio420X FPGA mezzanine card) to low cost (i.e. WiSpy) that can be used to experiment with spectrum sensing. This hardware is available for use in several testbeds across the world (i.e. ORBIT, w-iLab.t, TWIST and LOG-a-TEC). Each testbed provides a specific mechanism to define, deploy and execute experiments making it difficult for an individual researcher to use more than one testbed. In this work we propose an information model for describing spectrum sensing functionality with the ultimate goal of developing and promoting a Common CR language that can describe the resources in existing GENI and FIRE testbeds.

We propose a framework for dynamic composition of communication services that is well suited for facilitating research and prototyping on real experimental infrastructures of remotely configurable embedded devices. By using the concept of composability, our framework supports modular component development for various networking functions, thereby promoting code reuse. The framework consists of four components: the physical testbed, the module library, the declarative language, and the workbench. Its reference implementation, ProtoStack, developed using semantic web technologies, supports remote experimentation on sensor platform-based infrastructure and is thus well suited also for experimenters who do not possess their own physical experimentation infrastructure. We illustrate how ProtoStack supports research in service-oriented networks and cognitive networking. The cost of increased flexibility and prototyping speed of the protocol stack is paid in terms of increased memory footprint, processing speed, and energy consumption. Compared to the most related noncomposable approach, the CRime library used by ProtoStack has a 16--17% larger footprint, takes 2.4 times longer to execute an open-send-recv-close sequence, and consumes 1.6% more power in doing so. Even though with ProtoStack more resources are consumed by the node, the tradeoff in terms of prototyping speed pays off.

In this paper we propose a methodology for the experimental evaluation of a simple yet efficient power allocation game on a real-world outdoor experimental testbed. We adapt the existing theoretical framework and the ProActive Power Update (PAPU) algorithm to suit the constraints imposed by the LOG-a-TEC low-cost reconfigurable testbed. The resulting framework is implemented and evaluated in the ISM 2.4 GHz band. We study the effects of the empirical parameter estimation on the best response and players' strategies which represent the Nash equilibra. Our results show that for a certain cost range, the system can reach Nash equilibria. The equilibria and the convergence time are strongly influenced by each player's cost but also by the channel gains.