Inductive wireless charging of implantable medical devices necessitates careful control of magnetic and electric field emissions to meet strict safety regulations while delivering sufficient power. When designing a comfortable wireless charger that can operate over distances ranging to 10cm or more, it is difficult not to exceed the most stringent E-field limit of 83 V/m. This paper investigates electric field attenuation techniques for mid-range wireless power transfer at 6.78 MHz. Using feaFEAfinite element analysisfea like Ansys ® HFSS ™, three mitigation strategies are evaluated; (1) a high-permittivity dielectric shielding layer to absorb and redistribute electric fields, (2) multiple resonant tuning capacitors distributed along the transmitter coil to lower the voltage swing and confine high E-field regions, and (3) alternative coil-array transmitter topologies to spatially localize more confined E-fields. The results show that each technique significantly reduces the E-field magnitude without substantially affecting the H-field. Shielding the transmit coil attenuates the peak E-field from its initial 1416 V/m to 496 V/m, approximately a 65% reduction. Distributing the tuning capacitance into sixteen smaller capacitors yields a drop from the 1416 V/m to 231 V/m, approximately a 84% reduction. Both techniques preserve the required 8 A/m magnetic field. The third technique, a two-by-two coil array transmitter reduced the E-field from its 1416 V/m to 990 V/m (around 30% reduction), though with a slight magnetic field redistribution. All three methods combined, the E-field was successfully attenuated to 82 V/m, just below the strictest limit, without compromising power transfer efficiency. This research demonstrates a feasible approach and framework to safely extend the application of wireless charging for medical implants.

Continued progress towards energy-neutral Internet of Things (IoT) nodes expose the wireless communication link as the dominant energy bottleneck. While low-power wide-area network (LPWAN) technologies achieve long-range communication with multiple years of battery life, their active radios hinder reaching full energy neutrality. Long-range backscatter communication emerged as a key enabler, reaching one to three order of magnitude lower power consumption. New advancements leverage concepts from active radio systems such as chirp spread spectrum (CSS) modulation and integrate them on a low-power backscatter tag. This paper presents a comprehensive survey of long-range backscatter communication, using a bottom-up analysis spanning system topologies, hardware architecture, modulation techniques and medium access. Backscatter communication requires different topologies compared to active radios to reach longer communication distances. Different hardware architectures support backscattering a modulated signal with differing complexity, power consumption and spectral efficiency. At the physical layer binary switch-based modulation are well known and provide an easy form of modulation while chirp spread spectrum (CSS)-based modulation gain traction due to their robustness. Medium Access Control (MAC) techniques are examined with a focus on synchronization, concurrency and lightweight feedback mechanisms requiring low-power, low-complexity hardware. Building on these established solutions the paper evaluates the feasibility of long-range backscatter communication in different energy-neutral Internet of Things (IoT) applications. Starting from the available energy budget, harvested through solar, radio frequency (RF) or capacitive harvesting, feasible hardware, modulation and Medium Access Control (MAC) solutions are explored.

The deployment of Wireless Sensor Networks (WSNs) remains challenging and time consuming due to the manual commissioning, configuration, and maintenance of resource-constrained Internet of Things (IoT) devices. Achieving precise network-wide time synchronization in such systems further increases this deployment complexity. This paper presents a novel Real-Time Operating System (RTOS)-integrated time synchronization method that distributes an absolute Coordinated Universal Time (UTC) reference across the network using a single Global Navigation Satellite System (GNSS)-enabled host. The method extends the semantics of the RTOS tick count by directly linking it to a global time reference. Consequently, sensor nodes obtain a notion of UTC time and can execute time-critical tasks at precisely defined moments without requiring a dedicated Real-Time Clock (RTC) or GNSS module on each sensor node. This design reduces both hardware cost and overall system complexity. Experimental results obtained on custom-developed hardware running FreeRTOS demonstrate a task synchronization error below ±30 μs between the GNSS reference and a sensor node operating at a clock frequency of 32 MHz. Such precise network-wide synchronization enables more efficient channel utilization, reduces power consumption, and improves the accuracy of both local and coordinated task execution across multiple devices in WSNs. It therefore serves as a key enabler for self-deployable WSNs.

What is the optimal base station (BS) resource allocation strategy given a measurement-based power consumption model and a fixed target user rate? Rush-to-sleep in time, rush-to-mute in space, awake-but-whisper in power, or a combination of them? We propose in this paper an efficient solution to the problem of finding the optimal number of active time slots, active antennas, and transmit power at active antennas in a multiple-input multiple-output (MIMO) orthogonal frequencydivision multiplexing (OFDM) system under per-user rate and per-antenna transmit power constraints. The use of a parametric power consumption model validated on operator measurements of 4G and 5G BSs enhances the interpretation of the results. We discuss the optimal energy-saving strategy at different network loads for three BS configurations. Using as few BS antennas as possible is close to optimal in BSs not implementing time-domain power savings such as micro-discontinuous transmission ( mu DTX ). Energy-saving schemes that jointly operate in the three domains are instead optimal when the BS hardware implements timedomain power-saving modes, with a tendency for rush-to-mute in massive MIMO and for rush-to-sleep in BSs with fewer antennas. Median energy savings up to 30% and energy efficiency improvements up to 50% are achieved at low network loads.

Increasing attention is given to the upper mid-band or Frequency Range 3 (FR3), from 7 to 24 GHz, in the research towards sixth-generation (6G) networks. Promises of offering large data rates at favorable propagation conditions are leading to novel FR3 base station (BS) architectures, with up to thousands of antenna elements and radio-frequency (RF) chains. This work investigates the power consumption of prospective FR3 BSs and its relation to the delivered data rates. We model the power consumed by digital and analog signal processing, power amplifiers (PAs), and supply and cooling during four phases (data, signaling, micro-sleep, and idle) in downlink and uplink. Hybrid partially-connected beamforming is compared to fully-digital one. Results show that, for BS arrays with $1024$ antennas at $30\%$ of load, the PA consumes most of the power when $64$ or less RF chains are utilized, while the digital and analog processing consumption takes over when the number of RF chains is $512$ or more. The digital plus analog processing consumes $2\times$ to $4\times$ more than the PA for fully-digital beamforming. Hybrid beamforming achieves $1.3$ Gbit/s/user in downlink while improving the energy efficiency by $1.4\times$ compared to fully-digital beamforming.

In large-antenna systems, operating power amplifiers (PAs) close to saturation is key to achieving high energy efficiency. However, this operating regime inevitably introduces nonlinearities, whose resulting distortion can coherently combine at the user location and degrade system performance. Distortion-aware precoding techniques have been proposed to send this distortion in non-user directions. These distortion-aware precoders are typically designed under the assumption of a frequency flat channel. In contrast, we assume a multi-carrier, single-user downlink system that experiences frequency-selectivity. In multi-carrier systems, the non-linearity of the PA produces intermodulation products, creating inter-carrier interference so that the precoding problem becomes coupled across subcarriers. To tackle this, we propose a graph neural network (GNN)-based joint precoding method that optimizes precoding vectors across all subcarriers while accounting for distortion. Simulation results indicate that this joint GNN precoder achieves stable rates across a wide range of channel conditions, outperforming classical maximum ratio transmission and per-subcarrier GNN approaches, albeit at increased computational complexity. Additionally, we show that the need for such a method depends strongly on the channel conditions. As such, we propose a convolutional neural network-based classifier that selects the appropriate precoding method based on the current channel conditions, while balancing the achievable rate and complexity.

Data-driven Indoor positioning systems (IPS) commonly require ground truth (GT) labels and are prone to changing environments. This makes deployment labour-intensive and difficult to scale. This work proposes a label-efficient and selfsupervised ultrasonic IPS that removes these requirements by learning the spatial relationships of the environment directly from received channel observations. The method builds on channel charting (CC), using a triplet-based training objective and a graph neural network (GNN) to generate a latent chart that preserves geometric neighbourhood. An affine alignment step maps the learned chart to physical space using only a minimal number of reference points. The system is implemented and evaluated in an indoor testbed. Results demonstrate that the CC-based approach achieves positioning accuracy close to conventional supervised learning while requiring neither GT labels nor anchor coordinate knowledge, highlighting its potential for effortless and scalable deployment. Additional learning during operation mitigates the difficulties arising from a dynamically changing environment.

A plant's movements reflect its response to the environment, with suboptimal conditions causing rapid changes that indicate stress. Tracking plant movements presents a distinct way of assessing their physiological condition and detecting stress. This article introduces a novel autonomous wirelessly connected solution for plant movement detection based on a tailored design of a plant-wearable sensor system using an inertial measurement unit (IMU). The system supports two sensor configurations, a petiole and a leaf module, offering flexibility for a wide range of applications tailored to different crop species, cultivation practices, and operational requirements. Both designs minimize weight and power consumption, balancing this tradeoff depending on plant organ rigidity. This work focuses on the energy management and validation of the petiole sensor with a total weight of 15.70 g without casing and a lifetime of 15 months, supporting complete long crop cycles without battery recharging or replacement. The wireless capability allows straightforward deployment and real-time data transfer. Initial in situ validation confirms that our nonintrusive, cost-effective IMU-enhanced sensor system offers a promising approach for plant sensing and Internet of Things (IoT)-enabled crop monitoring in real-world conditions. The dual-sensor approach enables researchers and growers to adapt monitoring strategies based on specific needs such as battery life constraints, sensor weight limitations, and plant morphology, making it a versatile solution for precision agriculture.

Wireless power transfer (WPT) is a promising service for the Internet of Things (IoT), providing a cost-effective and sustainable solution to deploy so-called energy-neutral devices on a massive scale. The power received at the device side from a conventional transmit antenna with a physically small aperture decays rapidly with the distance. New opportunities arise from the transition from conventional far-field beamforming to near-field beam focusing. We argue that a physically large aperture, that is large with respect to the distance to the receiver, enables a power budget that remains practically independent of distance. Distance-dependent array gain patterns allow focusing the power density maximum precisely at the device location, while reducing the power density near the infrastructure. Physical aperture size is a key resource in enabling efficient yet regulatory-compliant WPT. We use real-world measurements to demonstrate that a regulatory-compliant system operating at sub-10 GHz frequencies can increase the power received at the device into the milliwatt range. Our empirical demonstration shows that power-optimal near-field beam focusing inherently exploits multipath propagation, yielding both increased WPT efficiency and improved human exposure safety.

The sixth generation of mobile networks (6G) can play a central role in shaping a sustainable future, the most compelling contemporary challenge. Connecting the unconnected, reducing carbon emissions of vertical sectors, and allowing heterogeneous types of intelligence (including humans) to safely and constructively interact in complex environments, are only a few of the several challenges that can be supported by 6G. However, this requires a careful design that balances positive and negative impacts of 6G, towards a sustainable and sustainability-enabling technology. This article presents a holistic view that translates the complex interplay between the 6G enabling effects and the sustainability of 6G by design, into concrete trade-offs and research questions. Starting from today's challenges for society and associated key values, we unfold the dilemma into a set of technical trade-offs, whose solutions span from technological innovations to standardization actions towards applicability.

When energy harvesting is not feasible or fails to provide sufficient power, the energy buffer of battery-powered Internet of Things (IoT) devices inevitably depletes. The proper disposal and/or replacement of depleted and end-of-life (EoL) batteries is challenging, especially in rural IoT deployments, where human intervention is cumbersome. When batteries are left in nature, they can pose a significant environmental risk, leaking harmful chemicals into the soil. This work proposes a novel contactless battery solution for longevity and recyclability, providing automated battery replacement using a short-range wireless power transfer (WPT) link instead of a direct battery-to-IoT node contact-based connection for powering the IoT device. It facilitates battery recovery at EoL by, e.g., an unmanned vehicle (UV), reducing the need for manual intervention. Unlike complex mechanical solutions or contacts prone to corrosion, a contactless approach enables easy replacement and improves reliability and longevity in harsh environments. A technical challenge is the need for an efficient contactless solution to enable the IoT node to get energy from the battery. This work elaborates an efficient wireless connection between the battery and IoT node, which ensures robustness in harsh environments. In addition, it examines the sustainability aspects of this approach. The WPT system is applied in two IoT node applications: polling-based and interrupt-based systems. The proposed solution achieves a transmitter-to-receiver efficiency of 72% and has an additional environmental impact of 2.34 kgCO2eq. However, its key advantage is the ease of battery replacement, which could significantly reduce the expected long-term environmental impact.

Massive MIMO systems are typically designed assuming linear power amplifiers (PAs). However, PAs are most energy efficient close to saturation, where non-linear distortion arises. For conventional precoders, this distortion can coherently combine at user locations, limiting performance. We propose a graph neural network (GNN) to learn a mapping between channel and precoding matrices, which maximizes the sum rate affected by non-linear distortion, using a high-order polynomial PA model. In the distortion-limited regime, this GNN-based precoder outperforms zero forcing (ZF), ZF plus digital pre-distortion (DPD) and the distortion-aware beamforming (DAB) precoder from the state-of-the-art. At an input back-off of -3 dB the proposed precoder compared to ZF increases the sum rate by 8.60 and 8.84 bits/channel use for two and four users respectively. Radiation patterns show that these gains are achieved by transmitting the non-linear distortion in non-user directions. In the four user-case, for a fixed sum rate, the total consumed power (PA and processing) of the GNN precoder is 3.24 and 1.44 times lower compared to ZF and ZF plus DPD respectively. A complexity analysis shows six orders of magnitude reduction compared to DAB precoding. This opens perspectives to operate PAs closer to saturation, which drastically increases their energy efficiency.

In [1], reference [30] is added and provided as follows.

With increasing interest in indoor positioning systems across various domains, such as industry, retail, and healthcare, the search for optimal solutions to meet the needs of different applications has gained significant momentum. This work highlights the potential of hybrid RF-acoustic systems combined with advanced machine learning models for robust, scalable, and energy-efficient indoor localization. The focus is on enhancing positioning algorithms for energy-neutral devices to improve accuracy, precision, reliability, and ease of installation. Traditional model-based (MB) methods, relying on line-of-sight (LoS) components, often struggle in challenging nonline-of-sight (NLoS) and reverberant environments. To address this, we propose data-driven neural network (NN) approaches capable of harnessing multipath components (MPCs) as additional information. The echoes in the room are exploited to improve accuracy. Various NN architectures, including multilayer perceptrons, (circular) convolutional neural networks, and graph neural networks (GNNs) are evaluated, in first instance using synthetic data. Results demonstrate that especially GNNs outperform MB methods, achieving superior accuracy in both LoS and NLoS scenarios. During the second phase, extensive real-life experiments are carried out. The GNN is evaluated using cross-validation, training on measurement data, and transfer learning (TL) within a reverberant NLoS environment. The cross-validation and TL demonstrate the practical feasibility. We report over 80% of improvement in 3-D positioning error compared to the MB technique.

Precision agriculture demands non-invasive, energy-efficient, and sustainable plant monitoring solutions. In this work, we present the design and implementation of a lightweight, batteryless plant movement sensor powered solely by RF energy. This sensor targets Controlled Environment Agriculture (CEA) and utilizes inertial measurements units (IMUs) to monitor leaf motion, which correlates with plant physiological responses to environmental stress. By eliminating the battery, we reduce the ecological footprint, weight, and maintenance requirements, transitioning from lifetime-based to operation-based energy storage. Our design minimizes circuit complexity while enabling flexible, adaptive readout scheduling based on energy availability and sensor data. We detail the energy requirements, RF power transfer considerations, integration constraints, and outline future directions, including multi-antenna power delivery and networked sensor synchronization.

This paper addresses the optimized base station (BS) resource allocation strategy in a massive multiple-input multiple-output (MIMO)-orthogonal frequency-division multiplexing (OFDM) system, aiming to minimize energy usage while meeting fixed downlink user data rates. We explore whether it is better to save energy by minimizing the active time slots ("rush-to-sleep"), the active antennas ("rush-to-mute"), the transmit power ("awake-but-whisper"), or combining these approaches. We utilize a measurement-based parametric power consumption model of sub-6 GHz BSs. We show that the formulated problem can be optimally solved by exploiting its convexity. The performance analysis across different network loads suggests that a rush-to-mute is close-to-optimal at most network loads when the BS hardware does not include time-domain power-saving modes like micro-discontinuous transmission (mu DTX). Median energy savings of 24% are achieved over the rush-to-sleep and awake-but-whisper at medium network loads. With enabled time-domain hardware power-saving modes, operating in the three energy-saving domains is the optimal strategy. The average power consumption decreases and median energy savings against the three specific schemes reach 18% at medium network loads.

Wireless power transfer (WPT) technologies hold promise for enhancing device autonomy, particularly for energy-limited IoT systems. This paper presents experimental results on coherent and non-coherent transmit diversity approaches for WPT, tested in the near field using the Techtile testbed. We demonstrate that a fully synchronized beamfocusing system achieves a 14 dB gain over non-coherent transmission, consistent with the theoretical 14.9 dB gain for a 31-element array. Additionally, phase alignment errors below 20 degrees result in less than 1 dB of gain loss, while errors exceeding 40 degrees lead to losses over 3 dB. These findings suggest that phase coherency requirements for WPT can be relaxed, and that scaling the number of antennas is a promising strategy for improving power transfer efficiency.

The vast bandwidth available at sub-THz frequencies holds great promise for high-speed wireless access, precise localization, and advanced sensing applications. However, fundamental physical constraints and technological limitations make the deployment of reliable sub-THz networks challenging. We propose a new paradigm for sub-THz coverage by transmitting the RF signals over polymer microwave fibers (PMFs) that interconnect low-complexity radio units (RUs) in a daisy-chain configuration. The distributed architecture ensures that user equipments (UEs) connect to RUs in their proximity, reducing path loss and mitigating blocking. The RUs leverage low-complexity, compact integrated antenna modules. Additionally, dual-frequency tandem operation is proposed, integrating the sub-THz system with a sub-10 GHz system that provides control signalling and a robust fallback solution for the sub-THz system. This proposed tandem architecture can open up the full potential of sub-THz technology and paves the way to cost- and energy-efficient, high-performance, real-time connectivity in dynamic environments.

Forklifts are essential for transporting goods in industrial environments. These machines face wear and tear during field operations, along with rough terrain, tight spaces and complex handling scenarios. This increases the likelihood of unintended impacts, such as collisions with goods, infrastructure, or other machinery. In addition, deliberate misuse has been stated, compromising safety and equipment integrity. This paper presents a low-cost and low-power impact detection system based on multiple wireless sensor nodes measuring 3D accelerations. These were deployed in a measurement campaign covering realworld operational scenarios. An algorithm was developed, based on this collected data, to differentiate high-impact events from normal usage and to localize detected collisions on the forklift. The solution successfully detects and localizes impacts, while maintaining low power consumption, enabling reliable forklift monitoring with multi-year sensor autonomy.

Massive MIMO systems are moving toward increased numbers of radio frequency chains, higher carrier frequencies and larger bandwidths. As such, digital-to-analog converters (DACs) are becoming a bottleneck in terms of hardware complexity and power consumption. In this work, non-linear precoding for coarsely quantized downlink massive MIMO is studied. Given the NP-hard nature of this problem, a graph neural network (GNN) is proposed that directly outputs the precoded quantized vector based on the channel matrix and the intended transmit symbols. The model is trained in a self-supervised manner, by directly maximizing the achievable rate. To overcome the non-differentiability of the objective function, introduced due to the non-differentiable DAC functions, a straight-through Gumbel-softmax estimation of the gradient is proposed. The proposed method achieves a significant increase in achievable sum rate under coarse quantization. For instance, in the single-user case, the proposed method can achieve the same sum rate as maximum ratio transmission (MRT) by using one-bit DACs as compared to 3 bits for MRT. This reduces the DAC's power consumption by a factor 4-7 and 3 for baseband and RF DACs respectively. This, however, comes at the cost of increased digital signal processing power consumption. When accounting for this, the reduction in overall power consumption holds for a system bandwidth up to 3.5 MHz for baseband DACs, while the RF DACs can maintain a power reduction of 2.9 for higher bandwidths. Notably, indirect effects, which further reduce the power consumption, such as a reduced fronthaul consumption and reductions in other components, are not considered in this analysis.