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.

The use of a reliable and accurate localization system is essential to ensure the smooth integration of visually impaired people into society and making their daily lives easier. Many of the proposed systems are based on the use of Machine Learning (ML) where increasingly complex models are introduced. However, this is not always necessary given that comparable performance can be achieved with simpler models by better understanding and adjusting key model parameters that impact model performance. In this paper, we present an indoor localization framework that applies a simple 1D Convolutional Neural Network (CNN) on a real collected LoRa dataset. We concentrate our work on studying and analyzing the impact of essential parameters and data criterion on the efficiency and performance of the system. We mainly study the impact of the LoRa frequency, the impact of the distribution of training points and the impact of the quality of collected fingerprints. Several conclusions have been drawn showing the correlation between localization performance and the specific features of propagation and model training.

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.

Location information serves as the fundamental element for numerous Internet of Things (IoT) applications. Traditional indoor localization techniques often produce significant errors and raise privacy concerns due to centralized data collection. In response, Machine Learning (ML) techniques offer promising solutions by capturing indoor environment variations. However, they typically require central data aggregation, leading to privacy, bandwidth, and server reliability issues. To overcome these challenges, in this paper, we propose a Federated Learning (FL)-based approach for dynamic indoor localization using a Deep Neural Network (DNN) model. Experimental results show that FL has the nearby performance to Centralized Model (CL) while keeping the data privacy, bandwidth efficiency and server reliability. This research demonstrates that our proposed FL approach provides a viable solution for privacy-enhanced indoor localization, paving the way for advancements in secure and efficient indoor localization systems.

To localize objects in an indoor environment, several methods are used such as trilateration and fingerprinting. These methods are based on the Received Signal Strength Indicator (RSSI), which is very sensitive to propagation factors indoor due to the existence of different static and dynamic obstacles. The use of RSSI causes the problem of missing data because the RSSI transmitted by an access point is not correctly received by the object sensors. To deal with the problem of missing data, researchers propose several completion methods, nevertheless no definitive solution has been brought. Therefore, we propose in this paper to use Generative Adversarial Networks for data recovery in order to be used efficiently for objects localization. The results based on the simulation data show an improvement of the localization accuracy compared to the classical methods. The simulations were performed with 10% and 50% of missing data and improved of 29.19% and 37.51 % respectively the localization accuracy.

The proliferation of the Internet of Things fosters collaboration among connected devices for tasks like indoor localization. However, existing indoor localization solutions struggle with dynamic and harsh conditions, requiring extensive data collection and environment-specific calibration. These factors impede cooperation, scalability, and the utilization of prior research efforts. To address these challenges, we propose FeMLoc, a federated meta-learning (MTL) framework for localization. FeMLoc operates in two stages: 1) collaborative meta-training, where edge devices contribute diverse localization data to train a global meta-model using a combination of model-agnostic MTL and federated averaging techniques and 2) rapid adaptation for new environments, where the pretrained global meta-model initializes localization models, requiring only minimal fine-tuning with a small amount of new data. In this article, we provide a detailed technical overview of FeMLoc, highlighting its unique approach to privacy-preserving MTL in the context of indoor localization. Our performance evaluations on real-world data sets, including UJIIndoorLoc, demonstrate the superiority of FeMLoc over state-of-the-art methods, enabling swift adaptation to new indoor environments with reduced calibration effort. Specifically, FeMLoc achieves up to 80.95% improvement in localization accuracy compared to the conventional baseline neural network (NN) approach after only 100 gradient steps. Alternatively, for a target accuracy of around 5m, FeMLoc achieves the same level of accuracy up to 82.21% faster than the baseline NN approach. This translates to FeMLoc requiring fewer training iterations, thereby significantly reducing fingerprint data collection and calibration efforts. Moreover, FeMLoc exhibits enhanced scalability, making it well-suited for location-aware massive connectivity driven by emerging wireless communication technologies.

The proliferation of connected devices in indoor environments opens the floor to a myriad of indoor applications with positioning services as key enablers. However, as privacy issues and resource constraints arise, it becomes more challenging to design accurate positioning systems as required by most applications. To overcome the latter challenges, we present in this paper, a federated learning (FL) framework for hierarchical 3D indoor localization using a deep neural network. Indeed, we firstly shed light on the prominence of exploiting the hierarchy between floors and buildings in a multi-building and multi-floor indoor environment. Then, we propose an FL framework to train the designed hierarchical model. The performance evaluation shows that by adopting a hierarchical learning scheme, we can improve the localization accuracy by up to 24.06% compared to the non-hierarchical approach. We also obtain a building and floor prediction accuracy of 99.90% and 94.87% respectively. With the proposed FL framework, we can achieve a near-performance characteristic as of the central training with an increase of only 7.69% in the localization error. Moreover, the conducted scalability study reveals that the FL system accuracy is improved when more devices join the training.

Response to industrial 4.0, massive devices are supposed to be equipped to collect and process a large amount of data. The high-dynamic environment is therefore created because of changes and moves of various equipments. In addition, indoor visible-light localization has attracted wide attention because of the popularity of the light-emitting diode (LED). However, most researches focus on the visible-light localization in the static environment. In this paper, we propose a high resolution visible-light localization approach in the dynamic industrial environment. This approach is based on the Particle Swarm Optimization (PSO) algorithm because it can reduce the influence of changes in a dynamic environment, but, it still be influenced. Thus, the PSO localization further supported by a neural network (NN) to achieve high and robust localization accuracy. Simulations are conducted to verify that the proposed approach has less sensitivity to environment's changes and three times coverage with localization less than 0.1 m compared to the conventional PSO algorithm.

Indoor localization can be based on a matrix of pairwise distances between nodes to localize and reference nodes. This matrix is usually not complete, and its completion is subject to distance estimation errors as well as to the noise resulting from received signal strength indicator measurements. In this paper, we propose to use convolutional neural networks in order to denoise the completed matrix. A trilateration process is then applied on the recovered euclidean distance matrix (EDM) to locate an unknown node. This proposed approach is tested on a simulated environment, using a real propagation model based on measurements, and compared with the classical matrix completion approach, based on the adaptive moment estimation method, combined with trilateration. The simulation results show that our system outperforms the classical schemes in terms of EDM recovery and localization accuracy.

This paper presents a localization algorithm based on the identification of a linear expression connecting a vector of mean received powers to a vector of Cartesian coordinates. The linear expression is based on a Taylor expansion of the received power collected on a set of measurement points. The terms of the Taylor expansion are then integrated in a transfer matrix able to predict a vector of received power from a composite vector of Cartesian coordinates. Using the pseudo inverse of this matrix it is then possible to find the Cartesian coordinates of any unknown reception point, from the vector of mean received powers at this point.

This paper presents a comparison study of different similarity metrics used for RSSI fingerprint based indoor localization. These metrics are used for nearest neighbor search which is a crucial step in fingerprint localization system. Including Euclidean distance, Manhattan distance and Gauss distance, the present study compares the localization error respect to a proposed parameter named “error density”. This latter is related to the location error and the size of the studied area. In addition, different methods of combining the locations of neighbors have been introduced to estimate the current position and their performances have been compared. Extensive implementation details are discussed and simulation is conducted to compare them. Obtained results show that the Kernel method combined with the weighted average exhibits the localization accuracy on the studied dataset.