UWB TDoA localization accuracy degrades in industrial non-line-of-sight (NLOS) environments, where traditional methods of excluding NLOS links are often infeasible and degrade geometric precision. To address these limitations, we propose a novel position correction method using a transformer encoder. The model directly processes raw channel impulse responses (CIRs) from all available anchors by first partitioning them into patches. These patches are converted into tokens and combined with novel spatial positional encodings before the transformer learns their complex interdependencies to compute a final position correction. We analyze multiple patching and encoding strategies to evaluate their impact on performance and scalability. Based on experiments on real-world UWB measurements, our approach can provide accuracies of up to 0.39 m in a complex environment consisting of (almost) only NLOS signals, which is an improvement of 73.6% compared to the TDOA baseline.

This paper provides an overview of the domain challenges, use cases, objectives, high-level concepts, intended innovations, and expected impact of the DistriMuSe project. The project’s main aim is to enhance human health and safety by improved sensing of human presence, behaviour, intentions and vital signs in a collaborative or common environment by means of multi-sensor systems, distributed processing and machine learning. The use cases address challenges in health monitoring of elderly, sleep and exercise, of drivers and vulnerable road users in traffic and of people interacting with robots in a factory environment. Technical development in the project focuses on unobtrusive monitoring sensors, multi-sensor systems, distribution of computation and intelligence, and domain specific needs for the use cases.

Ultra-wideband (UWB) localization systems commonly rely on Two-Way Ranging (TWR) or Time Difference of Arrival (TDoA) techniques, each associated with its strengths and limitations. While TWR offers accurate ranging without clock synchronization, it requires frequent packet exchanges. In contrast, TDoA enables scalable, low-latency localization but suffers from synchronization and multipath-induced errors. Although many error correction approaches have been proposed for TWR, scientific work on error correction for TDoA is lacking. In this work, we propose a hybrid learning-based approach that leverages data collected from TWR systems to enhance time of arrival (ToA) error correction in TDoA-based localization. The motivation behind this strategy is: (1) labeled TWR datasets are more readily available in the scientific community, and (2) obtaining accurate error labels in real-life conditions for TWR is significantly easier than for TDoA. To this end, we train a convolutional neural network (CNN) using channel impulse response (CIR)s and their corresponding ranging error labels from a TWR system. Once trained, the model is applied directly, without retraining or fine-tuning to CIRs captured in a TDoA setup. Experimental results demonstrate that existing error correction approaches, such as anchor node selection, fail when encountering strong multipath environments, with 11.6% of positions not being calculated due to the removal of too many CIRs. In contrast, our approach enables significant gains in positioning accuracy in dense multipath environments, where only a limited number of signals are received.

This paper explores the use of a commercial-off-the-shelf impulse-radio ultra-wideband transceiver with 0.5 GHz bandwidth for ranging and respiration rate monitoring of a moving target. Respiration causes the chest to change in shape, which alters the target’s radar cross section, which causes periodical changes in the channel impulse response. A median filter, Viterbi algorithm and a particle filter are used to determine the range of the target, band-pass filtering and fourier transforming are used for respiration rate estimation. The results demonstrate that accurate localization and respiration rate monitoring is possible under controlled settings, but not in a realistic scenario.

Next-generation ultrawideband (UWB) applications require high-performance and fully integrated UWB antenna systems to guarantee accurate localization and sensing in challenging Internet of Things (IoT) environments. This article proposes an entire system-level simulation framework that accelerates the design and optimization of integrated UWB antenna systems for various IoT applications by accurately predicting the effects of the entire system integration environment and enabling time-efficient optimization of system-level metrics. These metrics include the system fidelity factor and the distance estimation error in full 3-D, which are required to minimize orientation-specific pulse distortion and phase-center variation. To reconcile fast and accurate system-level performance prediction with reduced computational resources, the UWB link is partitioned, enabling the combination of standalone full-wave antenna simulations, UWB front-end circuit models, and UWB wireless channel models considering all antenna and circuit imperfections. Moreover, this simulation framework is the first to include the Huygens’ field equivalence principle to efficiently and accurately model the entire system environment, crucial to ensure high-performance UWB antenna systems for integrated IoT applications. To validate the simulation framework, an extensive time-domain measurement campaign was performed on a representative UWB link, including multiple integration platforms. Simulation and measurement results correspond well and show that the presence of the actual integration platform significantly impacts the system performance along different orientations. The simulation framework is several orders of magnitude faster than what is currently achievable with conventional electromagnetic field simulators and facilitates the development of high-performance and fully integrated UWB systems that satisfy the needs of demanding IoT applications.

Driven by the increasing demand for data, connectivity and automation, the amount of Internet of Things (IoT) devices continues to expand across consumer electronics and industrial applications. Integrating energy harvesting (EH) technologies as a battery-free alternative requires a controlled setting to emulate realistic EH systems. Yet, current tools are often not sufficiently accurate to model all hardware and software components of an EH system. Moreover, it is hard to mimic realistic energy availability scenarios, which are crucial for the design and optimization of EH systems. Therefore, a novel emulation platform is introduced that facilitates the development, testing, and optimization of complete and realistic EH systems. This emulator device replicates the behavior of the entire EH system. It consists of a Raspberry Pi 5 with a custom developed add-on hardware hat and dedicated software. This add-on hardware hat provides a stable output voltage between 1.2 and 3.6 V to the IoT device under test and contains a current measurement circuit with an accuracy of$0.6~\mu $A. The software includes novel and accurate digital-twin models of the energy harvester, the storage element, and the power management unit, enabling the emulation of various EH scenarios. Two application scenarios are demonstrated, a wireless bluetooth low energy (BLE) heart rate sensor optimized for ambient light EH and a compact wireless BLE temperature sensor optimized for radio-frequency EH. The proposed emulation platform enables rapid EH design evaluation under varying energy conditions, streamlining development and validating system reliability. This could be an important step toward establishing EH as a sustainable and widely adopted alternative to conventional battery-powered systems.

This paper proposes two approaches to significantly improve the detection rate of abnormal heartbeats in an Electrocardiogram (ECG) based deep learning heartbeat classifier. We introduce an automated feature selection procedure using the Kendall rank correlation coefficient to improve the performance of already existing classifier models. Further, we propose a methodology to cope with the class imbalance present in many ECG and other medical datasets by using a weighted loss function. The proposed methods demonstrate a significant improvement in the detection of Supraventricular Ectopic Beat (SVEB) and Ventricular Ectopic Beat (VEB) type heartbeats. Boasting an impressive 20% increase in terms of recall for the SVEB class when compared to state-of-the-art classifiers. This advancement could lead to more reliable and efficient tools for early arrhythmia detection, particularly beneficial in places where professional medical care is not easily accessible.

Indoor positioning using UWB has gained popularity due to its low cost while still providing centimeter-level accuracy. Currently, Convolutional Neural Network (CNN)-based approaches are often proposed for NLOS detection, error correction, etc. to make these UWB positioning systems more accurate. Transformer (TF) networks have shown to be a more capable alternative in several other domains, but have not been used for UWB fingerprinting. We present two novel TF-based approaches: one that processes channel impulse responses (CIR) directly, and a second one that uses cross-attention to incorporate anchor position information to improve geometric understanding. Moreover, we propose a second innovation by combining fingerprinting with a time calibration method that synchronizes the CIR data using a TDOA-based setup. This second innovation can be used with our novel approach, or with other state-of-the-art fingerprinting methods. The proposed models are evaluated in an industrial environment and outperform previous state-of-the-art CNNs in both LOS and NLOS situations, reaching accuracies with errors as low as 3 cm in real-life conditions, while having lower complexity and requiring fewer samples.

Addressing long-range dependencies in blind co-channel interference waveforms typically requires convolutional networks with large kernels or significant depth, which are resource-intensive. This paper presents a streamlined UNet architecture integrated with fast Fourier convolution blocks and a long short-term memory in the bottleneck, designed to efficiently capture these dependencies. By leveraging the Fourier domain for global feature processing, our architecture reduces the model's complexity without compromising performance. Compared to the leading benchmark model (a deep UNet), our approach yields a 26.5% improvement in mean square error, while reducing multiply-accumulate operations and the number of model parameters by 76.8% and 76.3% respectively, demonstrating a significant enhancement in both accuracy and efficiency for interference cancellation in constrained computational environments.

Ultra wideband (UWB) is a popular technology to address the need for high-precision indoor positioning systems in challenging industry 4.0 use cases. In line-of-sight (LOS) environments, UWB positioning errors in the order of 1–10 cm can be achieved. However, in non-line-of-sight (NLOS) conditions, this precision drops significantly, with errors typically >30 cm. Machine learning (ML) has been proposed to improve the precision in such NLOS conditions, but is typically environment-specific and lacks generalization to new environments and UWB configurations. As such, it is necessary to collect large data sets to train a neural network (NN) for each new environment or UWB configuration. To remedy this, this article proposes automatic optimizations for transfer learning (TL) deep NNs toward new environments and UWB configurations. We analyze error correction and (N)LOS classification models, using either feature- or channel impulse response (CIR)-based input data. Our TL solutions show a 50% error improvement and 15% (N)LOS classification accuracy improvement (for both feature- and CIR-based approaches) compared to a model trained in a different environment. We also analyze the impact on TL using a limited number of samples (25 to 400 samples). The highest accuracy is typically achieved by the CIR-based approach, where with only 50 samples from the new mixed (N)LOS environment, we show ±10 cm precision after error correction with 93% (N)LOS detection. The presented results demonstrate high-precision UWB localization (from 643 to 245 mm) through ML with minimal data collection effort in challenging NLOS environments.

This paper proposes a fully unsupervised technology recognition method using deep clustering for identifying wireless technologies to learn from raw data without requiring manual label annotations. Identifying and recognizing wireless technologies is important to realize effective spectrum management. Recent scientific publications utilized supervised deep learning with great success to train AI models to recognize different wireless technologies based on labeled RF signal datasets. However, assigning labels to wireless technology signal datasets for supervised deep learning is time consuming and may not always be practical. To remedy this issue, the proposed method combines an autoencoder and clustering layers, extended with a Gaussian Mixture Model, to learn low-dimensional salient features from raw IQ data and cluster them into distinct wireless technologies. The optimal input dimensions and performance of the method in different signal-to-noise ratio conditions are analyzed for a set of several different low-power wide area network (LPWAN) technologies: Sigfox, LoRa, IEEE 802.11ah, and IEEE 802.15.4g. Extensive simulations show that the proposed method for wireless technology recognition exhibits a recognition accuracy exceeding 90%, surpassing the performance of the state-of-the-art deep clustering algorithms by 4 to 5% in terms of accuracy.

Industrial wireless sensor networks offer a viable alternative to wired solutions where there is a lack of suitable communication infrastructure. Among these networks, time slotted channel hopping (TSCH) emerges as a noteworthy choice due to its capacity for deterministic latency, heightened reliability, and low power consumption. Nonetheless, challenges arise from the energy-intensive joining procedure and inherent idle listening associated with TSCH, impeding the integration of battery-powered end devices. Therefore, this article introduces a novel 6TiSCH low-power node (6LPN) that supports ultralow-power operations. The proposed solution optimizes the energy-intensive joining procedure through reduced advertisement channels, optimal scanning time, and a delayed join, achieving a 90% reduction in energy consumption. Furthermore, idle listening is eliminated by queueing downlink traffic in a 6TiSCH friend node (6FN), ensuring an 87%–94% reduction in power consumption during operational mode while maintaining an average latency of queued frames of 7.62 s. By comparing the impact of the optimizations on three Internet of Things (IoT) hardware platforms, we demonstrate that optimal results are obtained for devices with low RX and idle transceiver current. Finally, our solution maintains full backward compatibility with 6TiSCH and does not introduce additional control traffic.

Dynamic Spectrum Sharing (DSS) is an enabler for a seamless transition from 4G Long Term Evolution (LTE) to 5G New Radio (NR) by utilizing existing LTE bands without static spectrum re-farming. In this paper, we propose a cross-band DSS scheme that utilizes the Multimedia Broadcast Multicast Service over a Single Frequency Network (MBSFN) feature of an LTE network and the Multicast Broadcast Service (MBS) feature of an NR network. The proposed DSS scheme utilizes LTE and NR resource controllers to assign muted MBSFN subframes on the LTE band and muted MBS subframes on the NR band based on traffic needs. In contrast to the state-of-the-art, the proposed DSS scheme does not require a coordination signaling channel between the LTE and NR networks. Instead, a machine learning-based Technology Recognition and Traffic Characterization (TRTC) system is used to identify and characterize traffic patterns. The LTE and NR resource controllers use the TRTC to sense the muted subframes and offload traffic accordingly. On average, the proposed DSS, as compared to static band configuration, improves the LTE throughput, NR throughput, LTE band spectrum utilization efficiency, and NR band spectrum utilization efficiency by 13.5%, 8.3%, 11.8%, and 20.7%, respectively.

Automatic distance detection is essential for industrial safety processes, allowing detection of persons in unsafe zones, detecting obstacles nearby autonomous vehicles, etc. ultr-awideband (UWB) radar is a recent upcoming technology that is suitable for low-cost, device-free distance estimations to persons and obstacles. The excellent time properties of UWB allow the detection of distinct propagation paths from sender to receiver and reflecting objects. Currently, existing work utilizes dedicated UWB radar hardware. This paper focuses on off-the-shelf DW1000 UWB transceivers which have lower cost but also lower time resolution of the received channel impulse response (CIR). We analyze three data processing methods (ICIR, UCIR, and ACIR). Next, we analyze two distance estimation approaches in a realistic industrial environment: a mean based and a variance based. We demonstrate that selecting the best combination of data processing and distance estimator is crucial, allowing the detection of metal objects up to 900 cm with a mean accuracy of 9 cm, or the detection of persons with a mean absolute error of only 44 cm.

A batteryless Internet of Things (IoT) offers a sustainable alternative to battery-powered IoT devices, which produce billions of dead batteries every year. Devices are instead powered by a small supercapacitor, which is recharged by a renewable energy source. However, since IoT devices are often characterized by intermittent periods of high energy consumption followed by periods of reduced activity, conventional average energy consumption models cannot be used to assess if IoT devices can be powered by energy harvesters. Therefore, this article presents an alternative feasibility evaluation approach that focuses on modeling the worst case periods with peak energy consumption and short idle times, which pose the highest constraints on the capacitor’s behavior. This approach simplifies the characterization of the wireless technology energy consumption as these worst case periods can be determined by a few parameters. The methodology is then applied to combinations of popular IoT technologies (LoRaWAN, BLE Mesh, and 6TiSCH) and energy sources (solar, kinetic, and radio frequency energy) for two common IoT use cases. We show that the proposed parameters can be successfully extracted with power measurements for different network configurations and that the Power Management Unit configuration has a nonnegligible impact on the communication requirements. Finally, we discuss how to apply the model to other technologies and other use cases.

Activity recognition models based on wearable devices are becoming increasingly popular. However, models that are trained and tested on the same players show a large bias and are not generalizable to previously unseen players. In this paper, we tackle the badminton stroke recognition problem from this perspective, comparing the performance of individual and generalized models based on an accelerometer and a gyroscope, and identifying which components of the solution can maximize the performance of generalized models. First, we describe a simple convolutional neural network trained to classify 7 types of stroke. Second, the model is extended in a hybrid way to identify two additional classes (movement and rest). Third, data augmentation is applied on the training set. Fourth, transfer learning is applied to use data from the test player to fine-tune the generalized model and attempt to reach the performance of an individual model. These models are evaluated on a dataset collected from amateur players, both in a controlled environment and in a match simulation. The results showed a large difference between the performance of individual and generalized models; however, the latter could be improved by increasing the number of players in the training set, by data augmentation, and by transfer learning, highlighting the necessity of larger datasets in this field.

Within the context of the Internet of Things (IoT), many applications require high-quality positioning services. As opposed to traditional technologies, the two most recent positioning solutions: 1) ultra-wideband (UWB) and 2) (unmodulated) visible light positioning [(u)VLP] are well suited to economically supply centimeter-to-decimeter level accuracy. This manuscript benchmarks the 2-D positioning performance of an 8-anchor asymmetric double-sided two-way ranging (aSDS-TWR) UWB system and a 15-LED frequency-division multiple access (FDMA) received signal strength (RSS) (u)VLP system in terms of feasibility and accuracy. With extensive experimental data, collected at two heights in a 8 m by 6 m open zone equipped with a precise ground-truth system, it is demonstrated that both visible light positioning (VLP) and UWB already attain median and 90thpercentile positioning errors in the order of 5 and 10 cm in line-of-sight (LOS) conditions. An approximately 20-cm median accuracy can be obtained with uVLP, whose main benefit is it being infrastructureless and thus very inexpensive. The accuracy degradation effects of non-LOS (NLOS) on UWB/(u)VLP are highlighted with four scenarios, each consisting of a different configuration of metallic closets. For the considered setup, in 2-D and with minimal tilt of the object to be tracked, VLP outscores UWB in NLOS conditions, while for LOS scenarios similar results are obtained.

With the ongoing push towards an automated Industry 4.0, data-driven intelligent algorithms are getting more attention. Warehouse operators have traditionally required human labor to identify and register their assets. Autonomous flying drones will help alleviate this task by flying through the warehouse and detecting assets. This can be done based on vision, requiring expensive and energy consuming hardware, limiting drone flight time. In contrast, we propose a solution using radio-frequency identification (RFID) tags and machine learned algorithms to localize assets, which does not require a well-lit environment and can be processed in an energy efficient way. Our machine learning model achieves a 92–93 % accuracy, even when the drone is flying at different heights than the assets. Additionally, the model is easily implementable on off-the-shelf and low-energy consuming embedded hardware. This data-driven solution can easily be retrained for different environments and allows cheap RFID-based horizontal localization of assets in warehouses of the future.

Ultra wideband positioning systems typically use techniques such as two way ranging (TWR) or time difference of arrival (TDoA) to calculate the position of mobile tags. TDoA techniques require the transmission of only a single packet by the mobile tag, thus providing better scalability, higher update rates and less energy consumption than TWR techniques. However, the TDoA performance degrades heavily when a subset of the anchors are in non-line-of-sight (NLOS) conditions with the tag or with each other. To remedy this, we propose and compare different algorithms to select a subset of anchor pairs before calculating the TDoA position in 3 different conditions: LOS conditions between all devices, NLOS conditions between tag and anchor nodes and NLOS conditions between anchors and between tag and anchor nodes. We use an experimental setup with 1 tag and 8 anchor nodes to compare the accuracy gains obtained by using both simple algorithms and more complex machine learning (ML) based algorithms applied on the channel impulse responses of anchor pairs. By selecting the best anchor combinations our algorithms can reduce the positioning error by 75% (assuming perfectly known ground truth), by 19% using realistically low complexity algorithms and by 38% for ML based algorithms.

Wireless Sensor Networks (WSNs) operating in unlicensed frequency bands or employing battery-less devices, require a Duty Cycle (DC) limit to ensure fair spectrum access or limit energy consumption. However, in multi-hop networks, it is up to the network protocol to ensure that all devices comply with such DC restrictions. We therefore developed a distributed DC adherence algorithm that limits the DC of all devices without introducing any additional packet overhead. This paper presents a brief description of the algorithm and evaluates its performance through simulation. Our results show that the algorithm can limit the DC of all devices to ensure no devices must switch off. Our algorithm therefore provides a solution for WSNs where nodes must operate below a DC limit.

The numerous applications of industrial automation have always posed many challenges for wireless connectivity. In the last decade, IEEE 802.15.4e time-slotted channel hopping (TSCH) networks have provided high reliability and low-power operation in such challenging industrial environments. Typically, TSCH networks employ one modulation at the physical layer and are thus limited by the characteristics of the chosen modulation in terms of, among others, data rate, reliability and energy efficiency. To tackle these limitations and to improve network performance and flexibility in those challenging industrial environments, this work explores the simultaneous use of multiple modulations in a TSCH network. Traditionally, TSCH relies on fixed-duration slots, large enough to send a packet of any size given the fixed data rate. In order to avoid wasting airtime when simultaneously using modulations with different data rates, we propose the concept of slot bonding. This allows the creation of different-sized bonded slots with a duration adapted to the data rate of each chosen modulation. To analyze the proposed slot bonding technique, we formally describe the TSCH slot bonding problem in terms of optimizing the packet delivery ratio while minimizing radio on time, with the inclusion of parent selection and interference avoidance. Afterward, we propose a genetic algorithm that allows us to implement the problem and find solutions heuristically. Finally, we provide insights into preferred parent selection and modulation configurations by using this heuristic approach during extensive simulation experimentation in which the scalability advantage of slot bonding over longer fixed-duration slots is also shown.

In this study, an energy-efficient self-organized framework for sub-channel allocation and power allocation is presented for ultra-dense small cell networks, which can operate in both licensed and unlicensed bands. In order to protect legacy WiFi devices (operating in unlicensed bands), we consider the Long-Term Evolution (LTE) operation in unlicensed bands based on Carrier Sense Adaptive Transmission (CSAT), in which 'ON' and 'OFF' duty cycle approach is utilized. On the other hand, there are severe interference management problems among small cells (operating in licensed and unlicensed bands) and between macro cells and small cells (operating in licensed bands) due to co-channel and ultra-dense deployment of small cells. This article proposes a self-organized optimization framework for the allocation of sub-channels and power levels by exploiting a non-cooperative game with the objective to maximize the energy efficiency of dual-access small cells without creating harmful impact on coexisting network entities including macro cell users, small cell users, and legacy WiFi devices. Simulation results show that the proposed scheme outperforms (6 and 11 percent) and (8 and 18 percent) the round-robin and the spectrum-efficient schemes, respectively, for two different small cell scenarios. In addition, it is shown that for less channel state information (CSI) estimation errors ς = 0.02, the maximum performance degradation of the proposed scheme is reasonably small (5.5 percent) as compared to the perfect CSI.

Recently, the operation of LTE in unlicensed bands has been proposed to cope with the ever-increasing mobile traffic demand. However, the deployment of LTE in such bands implies sharing spectrum with mature technologies such as Wi-Fi. Several studies have discussed this coexistence problem by suggesting that LTE implements different adaptation mechanisms that allow transmission possibilities to Wi-Fi. While such adaptation mechanisms exist, they still negatively impact Wi-Fi performance, mainly due to the lack of collaboration/coordination mechanisms that inform about the co-located networks' activities. In this paper, we propose a distributed spectrum management framework that enhances the performance of Wi-Fi, as a particular case, by detecting harmful co-located wireless networks and changes the Wi-Fi's operating central frequency to avoid them. The framework is based on a Convolutional Neural Network (CNN) that can identify different wireless technologies and provides spectrum usage statistics. Experiments were carried out in a real-life testbed, and the results show that Wi-Fi maintains its performance when using our framework. This translates in an increase of at least 40% on the overall throughput compared to a non-managed operation of Wi-Fi.

Low power wide area networks support the success of long range Internet of things applications such as agriculture, security, smart cities and homes. This enormous popularity, however, breeds new challenging problems as the wireless spectrum gets saturated which increases the probability of collisions and performance degradation. To this end, smart spectrum decisions are needed and will be supported by wireless technology recognition to allow the networks to dynamically adapt to the ever changing environment where fair co-existence with other wireless technologies becomes essential. In contrast to existing research that assesses technology recognition using machine learning on powerful graphics processing units, this work aims to propose a deep learning solution using convolutional neural networks, cheap software defined radios and efficient embedded platforms such as NVIDIA's Jetson Nano. More specifically, this paper presents low complexity near-real time multi-band sub-GHz technology recognition and supports a wide variety of technologies using multiple settings. Results show accuracies around 99%, which are comparable with state of the art solutions, while the classification time on a NVIDIA Jetson Nano remains small and offers real-time execution. These results will enable smart spectrum management without the need of expensive and high power consuming hardware.

For continuous, persistent and problem-free operation of Industrial Wireless Sensor Networks (IWSN), it is critical to have visibility and awareness into what is happening on the network at any one time. Especially, for the use cases with strong needs for deterministic and real-time network services with latency and reliability guarantees, it is vital to monitor network devices continuously to guarantee their functioning, detect and isolate relevant problems and verify if all system requirements are being met simultaneously. In this context, this article investigates a light-weight telemetry solution for IWSNs, which enables the collection of accurate and continuous flowbased telemetry information, while adding no overhead on the monitored packets. The proposed monitoring solution adopts the recent Alternate Marking Performance Monitoring (AMPM) concept and mainly targets measuring end-to-end and hopby-hop reliability and delay performance in critical application flows. Besides, the technical capabilities and characteristics of the proposed solution are evaluated via a real-life implementation and practical experiments, validating its suitability for IWSNs.

As the Internet of Things (IoT) continues to expand, the need to combine communication technologies to cope with the limitations of one another and to support more diverse requirements will proceed to increase. Consequently, we started to see IoT devices being equipped with multiple radio technologies to connect to different networks over time. However, the detection of the available radio technologies in an energy-efficient way for devices with limited battery capacity and processing power has not yet been investigated. As this is not a straightforward task, a novel approach in such heterogeneous networks is required. This article analyzes different low-power wide-area network technologies and how they can be integrated in such a heterogeneous system. Our contributions are threefold. First, an optimal protocol stack for a constrained device with access to multiple communication technologies is put forward to hide the underlying complexity for the application layer. Next, the architecture to hide the complexity of a heterogeneous network is presented. Finally, it is demonstrated how devices with limited processing power and battery capacity can have access to higher bandwidth networks combined with longer range networks and on top are able to save energy compared to their homogeneous counterparts, by measuring the impact of the novel vertical handover algorithm.

With the emergence of the Internet of Things (IoT) and Industry 4.0 concepts, industrial applications are going through a tremendous change that is imposing increasingly diverse and demanding network dynamics and requirements with a wider and more fine-grained scale. Therefore, there is a growing need for more flexible and reconfigurable industrial networking solutions complemented with powerful monitoring and management functionalities. In this sense, this paper presents a novel efficient network monitoring and telemetry solution for Industrial Wireless Sensor Networks mainly focusing on the 6TiSCH Network stack, a complete protocol stack for ultra-reliable ultra-low-power wireless mesh networks. The proposed monitoring solution creates a flexible and powerful in-band network telemetry design with minimized resource consumption and communication overhead while supporting a wide range of monitoring operations and strategies for dealing with various network scenarios and use cases. Besides, the technical capabilities and characteristics of the proposed solution are evaluated via a real-life implementation, practical and theoretical analysis. These experiments demonstrate that in-band telemetry can provide ultra-efficient network monitoring operations without any effect on the network behavior and performance, validating its suitability for Industrial Wireless Sensor Networks.

Fingerprinting indoor localization provides high positioning accuracy with low cost and easy deployment. Considering the unsatisfying precision of received signal strength (RSS)-based fingerprinting, hybrid metrics including angle-of-arrival (AoA) and time-of-flight (ToF), are incorporated to the RSS fingerprinting system. To evaluate the positioning performance of hybrid metrics, the closed-form Cramér-Rao lower bound (CRLB) is derived in this paper. The existence conditions of CRLBs, as well as the relationship of the CRLBs between single and hybrid metrics is revealed. Numerical results based on an office building scenario show that hybrid metrics greatly improve the positioning performance and the robustness to measured standard deviations compared to the single metric's case. Furthermore, hybrid schemes of the AoA/RSS/ToF metrics are also investigated, and simulations reveal that the scheme of AoA/ToF-supporting access points (AP) enhanced with single RSS-supporting APs achieves the best positioning accuracy among all hybrid schemes.

Indoor Positioning Systems (IPS) using ultra-wideband (UWB) are used in several application domains to optimize production processes and save expensive man hour costs. To deploy such a system, most solutions rely on an existing backbone network that is used for communication between the anchors and the Real Time Localization System (RTLS), which calculates the location. Our solution aims to be easy to install by using an IoT-standardized and low-power sub-GHz radio as backbone communication medium. Furthermore, using this low-power radio allows us to decrease the overall energy consumption of the anchors. In the demo we showcase that our solution does not require any wired connections and is a factor five more energy efficient than existing implementations.

This paper presents a Convolutional Neural Network (CNN) approach for classification of low power wide area network (LPWAN) technologies such as Sigfox, LoRA and IEEE 802.15.4g. Since the technologies operate in unlicensed sub-GHz bands, their transmissions can interfere with each other and significantly degrade their performance. This situation further intensifies when the network density increases which will be the case of future LPWANs. In this regard, it becomes essential to classify coexisting technologies so that the impact of interference can be minimized by making optimal spectrum decisions. State-of-the-art technology classification approaches use signal processing approaches for solving the task. However, such techniques are not scalable and require domain-expertise knowledge for developing new rules for each new technology. On the contrary, we present a CNN approach for classification which requires limited domain-expertise knowledge, and it can be scalable to any number of wireless technologies. We present and compare two CNN based classifiers named CNN based on in-phase and quadrature (IQ) and CNN based on Fast Fourier Transform (FFT). The results illustrate that CNN based on IQ achieves classification accuracy close to 97% similar to CNN based on FFT and thus, avoiding the need for performing FFT.

A number of industrial wireless technologies have emerged over the last decade, promising to replace the need for wires in a variety of use cases. Except for customized time division multiple access (TDMA)-based wireless technologies that can achieve ultralow latency over a very limited area, wireless communication generally has reliability and latency issues when it comes to industrial applications. Closed loop communication requires high reliability (over 99%), limited jitter and latency, which poses a challenge especially over a wide area measuring in hundreds of meters. Extended coverage is promised with the advent of sub-GHz technologies, one of them being IEEE 802.11ah which is the only one that offers sufficient data rate for frequent bidirectional communication. Thus, we evaluated IEEE 802.11ah for low-latency time-critical control loops. We propose the network setup for adjusting the network dynamics to that of control loops, enabling limited jitter and high reliability. We explore the scalability of IEEE 802.11ah network hosting both control loops and monitoring sensors that periodically transmit measurements. Assigning the control loop end-nodes to dedicated restricted access window (RAW) slot results in over 99.99% successful deliveries. Furthermore, interpacket delay is concentrated around the cycle-time in the following or preceding beacon interval in case the beacon interval is at least half the value of the shortest cycle-time. Adjusting the beacon interval to the fastest control loop in the network ensures latency requirements at the cost of maximum achievable throughput and energy consumption.

The new medium access method of IEEE 802.11ah, called restricted access window (RAW), divides stations into different groups, and only allows stations in the same group to access the channel simultaneously, in order to reduce collisions and thus achieve better performance (e.g., throughput). However, the existing station grouping strategies only support homogeneous scenarios where all stations use the same modulation and coding scheme (MCS) and packet size. A surrogate model is an efficient mathematical model that represents the behavior of a complex system, trained with a limited set of labeled input-output data samples. In this article, we present a surrogate model that can accurately predict RAW performance under a given RAW configuration in heterogeneous networks. Different from the homogeneous scenario, heterogeneous networks are defined by a large number of parameters, leading to an enormous design space, i.e., the order of 10(17) possible data points. This is too big to achieve feasible training convergence. In this article, we present a novel training methodology that leads to a new design space with highly reduced size, i.e., the order of 10(5) data points. The surrogate model converges when less than 6000 labeled data points are used for training, which is only a tiny portion of the whole design space. The results show that, the relative error between model prediction and simulation results is less than 0.1 for 95% of the data points, in the areas of the design space studied. Its low complexity and high precision make the proposed model a valuable tool to develop real-time RAW optimization algorithms for heterogeneous IEEE 802.11ah networks.

Indoor positioning systems (IPS) aim to track objects, people or assets with the highest possible accuracy. In literature, it has been shown that among radio-frequency based technologies, ultra-wideband (UWB) is capable of providing em-level accuracy. However, this technology is not yet mature when it comes to large-area coverage and multi-user support. To remedy this, this paper proposes a TDMA protocol for a large scale localization network with numerous simultaneously active users. To this end, a full system was designed that handles the scheduling of the transmissions, the synchronization of the fixed nodes and the roaming of the mobile nodes. Moreover, the performance of the system has been analyzed by simulations using OMNeT++ and INET framework. The simulation was improved with real life experiments and has been used to evaluate full TDMA/TDoA approach. Our solution improves the scalability to 88.3 % effective spectrum usage, compared to only 18.6% when using ALOHA, while mobile nodes are able to roam successfully in 90% of the handovers.

The Restricted Access Window (RAW) mechanism proposed by IEEE 802.11ah promises to address one of the major problems of the Internet of Things (IoT): high channel contention in large-scale densely deployed sensor networks. The RAW feature allows the Access Point (AP) to divide stations into different groups, with only the stations in the same group being allowed to access the channel simultaneously. Existing station grouping strategies only support homogeneous scenarios, where all sensor stations have the same fixed data transmission interval, modulation and coding scheme (MCS) and packet size. In this paper, we present two contributions to address this issue. First, a surrogate model that predicts RAW performance given specific network conditions and RAW configuration parameters. It is fast to train and can be solved in real-time. Second, the Model-Based RAW Optimization Algorithm (MoROA), which uses the surrogate model to determine the optimal RAW configuration in real-time, for heterogeneous stations and dynamic traffic. We compare the accuracy of our surrogate model to simulation results. Performance of MoROA is compared to existing RAW optimization algorithms and traditional 802.11 channel access methods. The results shows that the trained surrogate model can accurately predict RAW performance with a relative error less than 7% and 10% for 95% and 98% of the RAW configurations respectively. MoROA achieves a throughput up to twice as high as traditional 802.11 channel access functions in dense heterogeneous networks.

Experimental wireless network research is often very time consuming and requires knowledge of multiple experimentation platforms (JFED, OMF, etc.), thereby hindering innovation specially from non-testbed experts. To foster innovation, this paper presents an intuitive wireless experimentation using the Node-RED framework. Within the framework, drag and drop components are combined to set-up wireless experiments in simulation and testbed environments, configure network stack and execute series of experiments. Furthermore, the intuitiveness of the Node-RED framework is demonstrated by using drag and drop components to optimize multiple conflicting objectives in simulation and in a real-testbed, without requiring advanced testbed knowledge.

IEEE 802.11ah is a new Wi-Fi standard operating on unlicensed sub-GHz frequencies. It aims to provide long-range connectivity to Internet of Things (IoT) devices. The IEEE 802.11ah restricted access window (RAW) mechanism promises to increase throughput and energy efficiency in dense deployments by dividing stations into different RAW groups and allowing only one group to access the channel at a time. In this demo, we demonstrate the ability of the RAW mechanism to support a large number of densely deployed IoT stations with heterogeneous traffic requirements. Differentiated Quality of Service (QoS) is offered to a small set of high-throughput wireless cameras that coexist with thousands of best-effort sensor monitoring stations. The results are visualized in near real-time using our own developed IEEE 802.11ah visualizer running on top of the ns-3 event-based network simulator.

Since Global Positioning System technologies cannot be used indoors, a significant amount of research focuses on developing radio-frequency-based alternatives for indoor localization. Unfortunately, most of the suggested solutions for indoor localization consist of theoretical work or have been evaluated in non-industrial environments, typically office spaces. To evaluate the influence of industrial environments on localization accuracy, in this paper, a time-of-arrival (ToA) approach was used to determine the stationary locations of a robot inside the w-iLab.t II testbed, an open industrial-like environment containing several metal obstacles. The ToA method utilizes the measured propagation time of a radio wave between a sender and receiver to estimate their corresponding distance. This paper evaluates several industrial-related deployment aspects that influence location accuracy and describes how their negative impact can be reduced, resulting in an almost 50% accuracy improvement in industrial environments.

Wireless network deployments in industry often grow organically with new technologies added over time, among which many use the non-licensed spectrum to avoid licensing costs. As a result, technologies competing for the same spectrum end up deployed in the same area, causing coexistence problems to manifest themselves at a later stage. To avoid unexpected performance degradation, there is a need to evaluate the impact of additional wireless technologies on an existing network before the actual deployment. This paper proposes to simplify the impact assessment by emulating the signals of the potential wireless network with a single software-defined radio. To evaluate the emulator's performance, the impact of Bluetooth on Wi-Fi technology is considered as the reference scenario. A series of real-life experiments with configurable traffic load and network scale are conducted to estimate the impact of Bluetooth network on a Wi-Fi link, and the corresponding measurements are repeated with the emulated Bluetooth signals. To the best of the authors' knowledge, we are the first to propose such a solution, and it is shown that the use of our emulator gives a reliable indication of the expected impact at the location of the Wi-Fi link. As such, this paper provides an important step toward a simple, cost efficient, and reliable solution, to assess the impact of a wireless network prior to its deployment.

Wireless communication is an enabling technology for industrial automation. For mobile industrial devices operating in large areas, the performance of the wireless handover process is crucial. For the welfare of industrial processes short time communication outage must be ensured, especially for time-critical traffic. This paper assesses the handover performance for three industrial real-life use cases with different requirements. It covers handover performance under heavy interference, its impact on time-critical traffic and on broadcast traffic latency, followed by lessons learned and opportunities for further research.

This demonstration presents a user-friendly web-based platform that supports intuitive remote benchmarking of different indoor localization solutions. It reduces the barriers for experimental evaluation and fair comparison of their performance across a set of testing environments. The platform is aimed at addressing the limitations in the current praxis of publishing indoor localization research evaluated only in local, potentially biased environments. To this end, it provides a holistic support for the benchmarking process offering: (i) multiple pre-collected raw data-traces from different RF technologies, (ii) high-level interface to control remote wireless testbed facilities, and (iii) a set of tools for creating, storing, comparing and visualizing the performance results of multiple indoor localization solutions.

In this demo we demonstrate the functionalities of a novel experimentation framework, called WiSHFUL, that facilitates the prototyping and experimental validation of innovative solutions for heterogeneous wireless networks, including cross-technology coordination mechanisms. The framework supports a clean separation between the definition of the logic for optimizing the behaviors of wireless devices and the underlying device capabilities, by means of a unifying platform-independent control interface and programming model. The use of the framework is demonstrated through two representative use cases, where medium access is coordinated between IEEE-802.11 and IEEE-802.15.4 networks.

Wireless systems often need to optimize multiple conflicting objectives (low delay, high reliability, and low cost), which are difficult to fulfill simultaneously. In such cases, the wireless system exhibits multiple optimal operation points, referred to as the optimal Pareto front (OPF). However, due to the large number of parameter settings to be evaluated and the time-consuming nature of performing wireless experiments, it is typically not possible to identify the OPF by exhaustively evaluating all possible settings. Instead, for many use cases, an approximation is good enough. To this end, this paper applies a multi-objective surrogate-based optimization (MOSBO) toolbox to efficiently optimize wireless systems and approximate the OPF using a limited number of iterations. Moreover, a real Wi-Fi conferencing scenario is optimized that has two conflicting objectives (exposure and audio quality) and four configurable parameters (Tx-Power, Tx-Rate, Codec Bit-Rate, and Codec Frame-Length). The benefits of using the MOSBO approach for such a network problem is demonstrated by approximating the OPF using 94 iterations instead of requiring the exploration of 6528 different parameter combinations, while still dominating 96.58% of the complete design space.

A large amount of research focuses on experimentally optimizing the performance of wireless solutions. Finding the optimal performance settings typically requires investigating all possible combinations of design parameters, while the number of required experiments increases exponentially for each considered design parameter. The aim of this paper is to analyze the applicability of global optimization techniques to reduce the optimization time of wireless experimentation. In particular, the paper applies the Efficient Global Optimization (EGO) algorithm implemented in the SUrrogate MOdeling (SUMO) toolbox inside a wireless testbed. Moreover, to cope with the unpredictable nature of wireless testbeds, the paper applies an experiment outlier detection which monitors outside interference and verifies the validity of conducted experiments. The proposed techniques are implemented and evaluated in a wireless testbed using a realistic wireless conferencing scenario. The performance gain and experimentation time of a SUMO optimized experiment is compared against an exhaustively searched experiment. In our proof of concept, it is shown that the proposed SUMO optimizer reaches 99.79% of the global optimum performance while requiring 8.67 times less experiments compared to the exhaustive search experiment.

Indoor localization and navigation using wireless sensor networks is still a big challenge if expensive sensor nodes are not involved.Previous research has shown that in a sparse distributed sensor network the error distance is way too high.Even room accuracy can not be guaranteed.In this paper, an easy-to-use generic positioning framework is proposed, which allows users to plug in a single or multiple positioning algorithms.We illustrate the usability of the framework by discussing a new hybrid positioning solution.The combination of a weighted (range-based) and proximity (range-free) algorithm is made.Both solutions separately have an average error distance of 13.5m and 2.5m respectively.The latter result is quite accurate due to the fact that our testbeds are not sparse distributed.Our hybrid algorithm has an average error distance of 2.66m only using a selected set of nodes, simulating a sparse distributed sensor network.All our experiments have been executed in the iMinds testbed: namely at "de Zuiderpoort".These algorithms are also deployed in two real-life environments: "De Vooruit" and "De Vijvers".

In a modern conference room, various of video and audio devices are provided to ensure efficient communications. This is commonly referred to as a conferencing system. Compared to wired conferencing systems, wireless systems require less deployment effort, but may become unreliable when the selected radio spectrum is highly occupied. This demo focuses on improving the quality of service of a commercial wireless conferencing system using dynamic channel selection based on real-time spectrum sensing. The proposed solution is verified in a large-scale wireless testbed, and the result shows that the link of the conferencing system is indeed more robust against interference when cognitive solution is applied.

A large amount of research focuses on experimentally optimizing performance of wireless solutions. Finding the optimal performance settings typically requires investigating all possible combinations of design parameters, and as a result the number of required experiments increases exponentially for each considered design parameter. However, Efficient Global Optimization (EGO) algorithms overcome this limitation and arrive at the global optimum performance in a very short time compared to the exhaustive search technique. In this demo, we apply the SUrrogate MOdeling (SUMO) toolbox, an efficient implementation of EGO algorithms, in order to improve the experimentation time of a realistic wireless conference solution. By tuning a speaker's transmit power and channel parameters, the SUMO toolbox searches for an improved listeners' audio quality having minimum transmission exposure. Moreover, the SUMO experiment is compared to an exhaustive search experiment and it is found that SUMO reached 99.51\% of the global optimum performance while requiring 10 times less experiments.

In a future 'internet of things', an increasing number of every-day objects are connected with each other. Nowadays, connectivity between these devices is supported by assigning each device to an existing (wireless) network. However, these networks do not take into account the individual needs of these devices, even though all these devices are very different in terms of application requirements and hardware capabilities. Moreover, multiple existing networks are often configured independent from each other without any interaction. As an alternative, this paper proposes and discusses a methodology that more efficiently supports network cooperation between heterogeneous devices. The paper argues for autonomously created communities of similar devices, that are able to negotiate with different co-located communities to further optimize their network performance. Different communities engage in cooperation by activating network service, but only when the end result is beneficial for all involved communities. In this paper, the concepts and advantages of this approach are discussed. In addition, a methodology is explored that is able to realize these concepts. Finally, based on this methodology, possible network solutions are presented, remaining challenges are listed and future research opportunities are identified.

Wireless sensor networks consist of several hundredths of simple sensing devices, equipped with a radio. They are typically used for monitoring and automation purposes of large areas. Due to their simplicity, these networks quickly run out of energy, and often have problems regarding scalability and available bandwidth. To solve these issues, current research is mostly limited to the addition of extra sinks to the network, or the use of gateways to request sensor data over the Internet. In this paper, we explore how wireless sensor networks can be combined with wireless mesh networks to obtain a more optimized solution. The mesh network can be used to connect separate sensor networks, to connect sensor nodes with a monitoring platform, or as a scalable backbone for sensor to sensor communication. Additionally, we give an overview of the advantages and disadvantages of existing interconnection techniques between wireless and mesh networks, and propose several new interconnection strategies. Finally, we identify remaining challenges, upon which future research can be based.