The rapid expansion of the Internet of Things enables seamless collaboration among connected devices, making indoor localization a critical component of industrial automation. However, conventional localization techniques struggle in dynamic environments. These methods depend on extensive, frequent data collection and environment-specific calibration, limiting scalability, interoperability, and effective use of prior research while imposing significant energy penalties that hinder practical deployment. To address these challenges, we propose an energy-efficient transfer learning (TL) based approach for visible light communication based indoor localization. Our method leverages TL to tackle environmental variability, including lighting fluctuations and physical obstacles that typically degrade localization performance, while simultaneously reducing computational overhead and energy consumption compared to conventional approaches. We also introduce a novel model efficiency (ME) metric, designed to integrate localization accuracy, energy efficiency, data efficiency, and transfer gain into a single evaluative measure for comprehensive optimization. Evaluations on a real-world dataset collected from a BOSCH factory demonstrate that our model achieves a 47% improvement in localization accuracy compared to a conventional model that does not utilize TL, and achieves up to 67.5% ME with TL models, compared to only 10.4% in conventional models under high-noise conditions. These results underscore the potential of our approach to deliver a highly efficient and scalable indoor localization system suitable for energy-constrained industrial applications.

Indoor localization systems are critical for Industry-4.0 applications.Visible Light Communication (VLC)-based systems offer advantages such as immunity to electromagnetic interference, high accuracy, and energy efficiency, but their performance is hindered by environmental variability across deployment areas. This paper presents a transfer learning (TL) framework for VLC-based indoor localization that enhances cross-area robustness while reducing data requirements. Using real-world measurements from four distinct areas of a Bosch manufacturing facility, we show that fine-tuning a deep neural network (DNN) pre-trained on source area data with only 40% of target-area measurements recovers 97% of full-data performance, thereby reducing the amount of target data needed. We further introduce the Transfer Efficiency Score (TES), a composite metric that identifies the most effective source models for transfer without exhaustive evaluation. Guided by TES, the framework selects the best source model, yielding localization errors as low as 40.9 cm with success rates above 86%. Validation on an unseen area confirms its generalization capability, with a 3.28 cm mean error achieved using the TES-selected source model. Overall, the proposed framework provides a scalable and data-efficient pathway for industrial VLC localization, substantially reducing deployment costs and computational burden while maintaining high accuracy in diverse and dynamic environments.

Accurate indoor localization is crucial in industrial environments. Visible Light Communication (VLC) has emerged as a promising solution, offering high accuracy, energy efficiency, and minimal electromagnetic interference. However, VLC-based indoor localization faces challenges due to environmental variability, such as lighting fluctuations and obstacles. To address these challenges, we propose a Transfer Learning (TL)-based approach for VLC-based indoor localization. Using real-world data collected at a BOSCH factory, the TL framework integrates a deep neural network (DNN) to improve localization accuracy by 47 %, reduce energy consumption by 32 %, and decrease computational time by 40 % compared to the conventional models. The proposed solution is highly adaptable under varying environmental conditions and achieves similar accuracy with only 30 % of the dataset, making it a cost-efficient and scalable option for industrial applications in Industry 4.0.

Subnetworks are expected to enhance wireless pervasiveness for critical applications such as wireless control of plants, however, they are interference-limited due to their extreme density. This paper proposes a goal-oriented joint power and multiple sub-bands allocation policy for interference coordination in 6G in-factory subnetworks. Current methods for interference coordination in subnetworks only focus on optimizing communication metrics, such as the block error rate, without considering the goal of the controlled plants. This oversight often leads to inefficient allocation of the limited radio resources. To address this, we devise a novel decentralized inter-subnetwork interference coordination policy optimized using a Bayesian framework to ensure the long-term stability of the subnetwork-controlled plants. Our results show that the proposed decentralized method can support more than twice the density of subnetwork-controlled plants compared to centralized schemes that aim to minimize the block error rate while reducing execution complexity significantly.

The precise and accurate indoor positioning using cellular communication technology remains to be a prerequisite for several industrial applications, including the emergence of a new topic of Integrated Sensing and Communication (ISAC). However, the frequently occurring Non-Line-of-Sight (NLoS) conditions in a heavy multipath dominant industrial scenario challenge the wireless signal propagation, leading to abnormal estimation errors (outliers) in the signal measurements taken at the receiver. In this paper, we investigate the iterative positioning scheme that is robust to the outliers in the Time of Arrival (ToA) measurements. The Iteratively Reweighted Least Squares (IRLS) positioning scheme formulated on the Least Squares (LS) is implemented to reject the outlier measurements and reweight the available ToA samples based on their confidence. Our positioning scheme is validated under 5G frequency bands, including the C-band (3.7 GHz) and the mmWave-band (26.8 GHz) in a Ray-Tracing enabled industrial scenario with different emulation setups.

In this paper, we develop a novel power control solution for subnetworks-enabled distributed control systems in factory settings. We propose a channel-independent control-aware (CICA) policy based on the logistic model and learn the parameters using Bayesian optimization with a multi-objective tree-structured Parzen estimator. The objective is to minimize the control cost of the plants, measured as a finite horizon linear quadratic regulator cost. The proposed policy can be executed in a fully distributed manner and does not require cumbersome measurement of channel gain information, hence it is scalable for large-scale deployment of subnetworks for distributed control applications. With extensive numerical simulation and considering different densities of subnetworks, we show that the proposed method can achieve competitive stability performance and high availability for large-scale distributed control plants with limited radio resources.

In the context of Industry 4.0, cutting-edge technologies such as Computational Offloading and Machine Learning, particularly Reinforcement Learning (Q-Learning), demand exceptionally low latency. These technologies are essential for mission-critical applications. To meet these demands, Edge Computing (EC) has been integrated, deploying robust devices at the network's edge. However, latency issues persist within the industrial infrastructure due to the inherent limitations of existing network technologies. Despite significant progress in mobile communication, including Fourth Generation (4G), Long Term Evolution (LTE), LTE-Advanced (LTE-A), and wireless protocols such as Wireless Fidelity (Wi-Fi), the industrial environment necessitates superior communication technology with exceptional Key Performance Indicators (KPIs). In this regard, 5G technology is ideally suited to address these requirements. The effectiveness of Computational Offloading is not always guaranteed. Its advantages depend on the availability of system resources and network bandwidth. To ensure a successful offloading process, the decision-making agent must address four critical questions: when, what, where, and how to offload. In this paper, we focus on tackling the “When to offload?” question and investigate how Reinforcement Learning (RL) agents can provide solutions. Our evaluation is based on whether the offloading decision made by the RL agent results in a positive or negative gain for the selected application. To fortify our findings, we compare the accuracy of our RL agent's decisions with theoretical models and an Oracle-based approach. Furthermore, we assess the decision-making efficiency of the RL agent in both emulated and real-world industrial network environments.

Private fifth-generation (5G) networks are increasingly becoming the industry’s choice of wireless communication networks for accelerating production processes. In this context, the role of 5G in providing precise positioning services in indoor industrial scenarios has also been actively discussed. However, the achievable indoor positioning accuracy depends on the radio propagation conditions persisting in the scenario. In this paper, using a Ray-Tracing (RT) engine, we investigate the radio environment in C-band (3.775 GHz) as well as the mmWave-band (26.85 GHz) for a detailed 3D geometric model of the dense clutter industrial production hall under different emulation setups and categorize the dominant Non-Line-of-Sight (NLoS) MultiPath Components (MPCs). We then evaluate the achievable Observed Time Difference of Arrival (OTDoA) based positioning accuracy in the C-band and the mmWave-band by computing the position of User Equipment (UE) using only first-arriving MPCs.

In this paper, we present the ray tracing (RT) simulation in the 3D model of one highly dense clutter industrial hall, which is scanned by laser scanner and reconstructed based on accurate point cloud. The whole processing chain from the scanning of the physical environment to running the simulation is presented in detail. To validate the simulation results, the synthetic channel characteristics and large-scale parameters, including delay spread (DS), angular spread (AS) and path loss (PL), are compared with those obtained from channel sounding measurement in both LOS and NLOS cases, at 6.75 GHz, 30 GHz and 60 GHz. The simulation results show that some scatters are significant in all bands and may be well identified and tracked. This indicates that our target to generate a deterministic channel model or a hybrid channel model at multi-band for industrial scenario may be possible.

Micro Aerial Vehicle (MAV)-assisted localization of wireless nodes in a given area is especially important in post-disaster scenarios where no other communication infrastructure is accessibleDifferent from related work solely focusing on simulations, we have done a measurement campaign using a multirotor MAV to derive a realistic propagation model for our scenario. Based on this propagation model and a novel positioning approach, the performance of different trajectories is reevaluatedOur results show that in practice, the variance of the measurements is higher, than commonly expected in related work on MAV-assisted localization. As a result, the study of our developed trajectories show localization errors of 6-7m in a 400x400m2area. We also show that adaptive trajectories outperform state-of-the-art algorithms and are able to localize at least 90% of the nodes in under 20 minutes. Thus, they also outperform other algorithms with respect to the effectiveness of the trajectory.

A new level of factory automation demands processing vast amounts of data, complex orchestration of cyber-physical systems, and coordination of computation as well as communication resources in real-time. Virtualization and decentralized computation is becoming a de-facto solution for factory automation. Edge Computing (EC) is a promising approach to achieve the low latencies required by many industrial systems. It employs resource rich edge servers distributed within a factory that are placed close to end devices and assist them in executing computation intensive tasks and also in coordinating with each other. This paper discusses the requirements and challenges of EC for factory automation applications. In a distributed EC infrastructure, safe and timely operation of industrial applications require load balancing and mobility support and thus a seamless service migration between the edge servers. With the recent advances in virtualization and due to its advantages, virtual machine (VM) and container technologies are pavings its way into factory. Though containers have some distinctive advantages over VMs in EC, the service live migration has comparatively high downtime. This paper proposes a novel live migration scheme called redundancy migration that reduces the downtime by a factor of 1.8 compared to the stock migration in linux containers.

Future factory automation systems are expected to process vast amounts of data and orchestrate complex cyber-physical components. Edge Computing (EC) is a promising approach to address the requirements set by upcoming industrial systems. While EC caters to the computation requirements, it requires a solution to perform flexible network management of these computation resources. Software-Defined Networking (SDN) is a promising candidate to tackle such challenges. While most of the related work on EC and SDN focuses on multimedia or automotive applications, this paper presents the relevance of both paradigms for industrial applications. By introducing two most prominent industrial use cases, namely proactive system surveillance and intelligent technical assistance, this paper discusses the challenges involved and proposes a solution space for realizing these applications using the combination of EC and SDN. Furthermore, it presents future research directions regarding the combination of both paradigms in the context of factory automation.