Richard Demo Souza’s research while affiliated with University of Oulu and other places

In an age characterized by seamless interconnectivity, the quantity of Internet of Things (IoT) devices has experienced substantial growth in recent years, with projections indicating further expansion. In this context, Federated Learning (FL) plays an important role in the future of wireless communications, offering numerous advantages over traditional centralized learning approaches, including data privacy preservation, reduced bandwidth usage, improved accuracy, and customization. However, selecting an appropriate wireless protocol and data transmission method for FL is crucial. In this work, we adopt the multichannel ALOHA protocol due to its asynchronous nature and simple implementation compared to other protocols. This paper focuses on optimizing multichannel ALOHA communication within a hierarchical FL system by creating a Device-to-Device (D2D) clustering scheme, which enables a single base station to serve more devices and drastically reduces the achievable error.

We propose and evaluate age of information (AoI)-aware multiple access mechanisms for the Internet of Things (IoT) in multi-relay two-hop networks. The network considered comprises end devices (EDs) communicating with a set of relays in ALOHA fashion, with new information packets to be potentially transmitted every time slot. The relays, in turn, forward the collected packets to an access point (AP), the final destination of the information generated by the EDs. More specifically, in this work we investigate the performance of four age-aware algorithms that prioritize older packets to be transmitted, namely max-age matching (MAM), iterative max-age scheduling (IMAS), age-based delayed request (ABDR), and buffered ABDR (B-ABDR). The former two algorithms are adapted into the multi-relay setup from previous research, and achieve satisfactory average AoI and average peak AoI performance, at the expense of a significant amount of information exchange between the relays and the AP. The latter two algorithms are newly proposed to let relays decide which one(s) will transmit in a given time slot, requiring less signaling than the former algorithms. We provide an analytical formulation for the AoI lower bound performance, compare the performance of all algorithms in this set-up, and show that they approach the lower bound. The latter holds especially true for B-ABDR, which approaches the lower bound the most closely, tilting the scale in its favor, as it also requires far less signaling than MAM and IMAS.

Wireless Energy Transfer (WET) combined with Simultaneously Transmitting and Reflecting Reconfigurable Intelligent Surface (STAR-RIS) technology offers a promising approach to optimize the recharging of Internet of Things (IoT) devices. In this work, we propose the use of STAR-RIS in the WET context to enable efficient recharging of IoT devices, with the goal of minimizing the total system recharging time while ensuring that each IoT device meets its minimum energy requirement. The optimization is performed using the Particle Swarm Optimization (PSO) technique, including the beamforming configuration of the power beacon (PB) as well as the phase and amplitude coefficients of the STAR-RIS elements. We compare two STAR-RIS operating protocols: time switching (TS) and energy splitting (ES). Simulation results indicate that it is possible to charge devices efficiently using only statistical channel state information (S-CSI), even in the absence of direct link between the PB and the IoT devices.

Earlier studies and field tests have extensively investigated Long Range (LoRa) direct-tosatellite (DtS) communications, confirming the feasibility of integration with Low Earth Orbit (LEO) satellites. These works identify the Doppler effect as one of the primary challenges. Therefore, there is a need for a robust solution to mitigate the impact of this phenomenon in order to improve the performance of LoRa DtS communications in a LEO scenario. This paper addresses this shortcoming by developing a solution to pre-compensate the Doppler shift. Specifically, we propose a method that allows end devices to estimate and pre-compensate the Doppler shift before initiating an uplink transmission. This framework, which requires satellites to broadcast Doppler Beacons, ensures compatibility with existing LoRaWAN end devices without requiring any hardware modifications. We leverage data from real-world LoRa satellites empirical telemetry to validate our proposed method. We analytically study packet losses due to Doppler shift across different carrier frequencies, specifically 401.5 MHz, 868 MHz, and 2 GHz. Our analysis also considers different satellite orbital heights, specifically 200 km and 518 km, as well as channel bandwidths of 31.25 kHz, 62.5 kHz, and 125 kHz. Our results demonstrate that the proposed solution effectively precompensates for the Doppler shift and mitigates the packet losses, extending the passing satellites effective visibility window duration. We examine the maximum Doppler shift in the communication channel and the calculate required Doppler Beacon bandwidth for different orbital altitudes, minimum elevation angles, and carrier frequencies. This study also investigates how the proposed framework affects the battery lifetime of the end device, showing a marginal decrease of 2.5% compared to traditional LoRaWAN operation.

Massive machine-type communication (MTC) addresses new IoT use cases such as connected vehicles, smart agriculture, and factory automation. However, major concerns face massive MTC applications, such as data privacy, latency constraints, and communication overhead. Distributed learning, an emerging technology that enables edge users to train various machine learning (ML) models without sharing raw data, is very promising for overcoming such concerns. However, distrib-uted learning still faces some difficulties related to the amount of training data, accuracy, complexity, and dynamic wireless environment. Herein, we elucidate different distributed learning methodologies, their limitations, and case studies in the spirit of massive MTC. Specifically, the article showcases the pros of Federated Learning (FL) in the context of dense MTC-based handwritten digits classification. After that, we highlight the differences between F1 and Federated Distillation (FD) and proceed to split, transfer, and meta-learning. Then, we discuss multi-agent reinforcement learning (MARL) in a UAV swarm empowered smart agriculture scenario. Finally, we present some challenges and future directions for distributed learning research in the context of massive MTC.

We evaluate the performance of the LoRaWAN Long-Range Frequency Hopping Spread Spectrum (LR-FHSS) technique using a device-level probabilistic strategy for code rate and header replica allocation. Specifically, we investigate the effects of different header replica and code rate allocations at each end-device, guided by a probability distribution provided by the network server. As a benchmark, we compare the proposed strategy with the standardized LR-FHSS data rates DR8 and DR9. Our numerical results demonstrate that the proposed strategy consistently outperforms the DR8 and DR9 standard data rates across all considered scenarios. Notably, our findings reveal that the optimal distribution rarely includes data rate DR9, while data rate DR8 significantly contributes to the goodput and energy efficiency optimizations.

This work aims at improving indoor positioning systems (IPS) by integrating multiple radio frequency techniques, namely Received Signal Strength Indication (RSSI), Angle of Arrival (AoA), and a combination of both, within the Bluetooth Low Energy (BLE) 5.1 framework. While AoA stands out for its precision, low energy consumption, and cost-effectiveness, RSSI is characterized by its simplicity and widespread availability. By resorting to a database of real RSSI and AoA measurements from a BLE 5.1 target node in a 14×8-meter environment, our work employs the Kalman Filter (KF) to improve the accuracy of multilateration, AoA combined with RSSI, and AoA-only algorithms. Moreover, we consider one more step in our IPS where the aforementioned KF-filtered outputs are then fused through a track fusion model. Results demonstrate that the proposed scheme, which we refer to as Angle-RSSI Fusion Localization (ARFL), significantly improves localization accuracy compared to other techniques. In particular, it reduces up to 81.61% in the average position error when compared to multilateration with KF. This advanced IPS offers a cost-effective and precise solution suitable for various applications in industries such as healthcare, commerce, and logistics.

Most works evaluating the performance of Multi-User Multiple-Input Multiple-Output (MU-MIMO) systems consider Access Points (APs) with fixed antennas, that is, without any movement capability. Recently, the idea of APs with antenna arrays that are able to move have gained traction among the research community. Many works evaluate the communications performance of Movable Antenna Arrays (MAAs) that can move on the horizontal plane. However, they require a very bulky, complex and expensive movement system. In this work, we propose a simpler and cheaper alternative: the utilization of Rotary Antenna Arrays (RAA)s, i.e. antenna arrays that can rotate. We also analyze the performance of a system in which the array is able to both move and rotate. We focus on narrowband machine-type communications use cases in indoor scenarios, where multiple devices communicate simultaneously with the same AP in the uplink. The movements and/or rotations of the array are computed in order to maximize the mean per-user achievable spectral efficiency, based on estimates of the locations of the active devices and using particle swarm optimization. We adopt a spatially correlated Rician fading channel model, and evaluate the resulting optimized performance of the different setups in terms of mean per-user achievable spectral efficiencies. Our numerical results show that both the optimal rotations and movements of the arrays can provide substantial performance gains when the line-of-sight components of the channel vectors are strong. Moreover, the simpler RAAs can outperform the MAAs when their movement area is constrained.

Wired charging and the need for battery replacements are critical barriers to unlimited, scalable, and sustainable mobile connectivity, motivating the interest in radio frequency (RF) wireless power transfer (WPT) technology. However, the inherently low end-to-end power transfer efficiency (PTE) and health/safety-related apprehensions about the technology are critical obstacles. Indeed, RF-WPT implementation and operation require efficient and cautious strategies and protocols, especially when targeting high-power charging, which constitutes the scope of this work. Herein, we overview the main factors affecting the end-to-end PTE of RF-WPT systems and their multiplicative effect and interdependencies. Moreover, we discuss key electromagnetic field (EMF) exposure metrics, safety limits, and approaches for efficient and EMF-aware deployment and operation. Quantitatively, we show that nearfield RF charging may significantly reduce EMF exposure, and thus must be promoted. We also present our vision of a cyber-physical system for efficient and safe wireless charging, specify key components and their interrelation, and illustrate numerically the PTE attained by two modern low-power multi-antenna architectures in a simple setup. Throughout the article, we highlight the need for high end-to-end PTE architectures and charging protocols transparently complying with EMF exposure regulations, and outline relevant challenges and research directions. This work expands the vision and understanding of modern RF-WPT technology and constitutes a step toward making the technology attractive for worldwide commercial exploitation.