Next-generation autonomous and networked industrial systems (i.e., robots, vehicles, drones) have driven advances in ultra-reliable low-laten-cy communications (URLLC) computing. These multi-agent require fast, communication-efficient, distributed machine learning (ML) to provide mission-crit-ical control functionalities. Distributed ML techniques, including federated (FL), represent a mushrooming multidisciplinary research area weaving together sensing, communication, learning. FL enables...

Wireless propagation is conventionally considered as the enabling tool for transporting information in digital communications. However, recent research has shown that perturbations of same electromagnetic (EM) fields are adopted data transmission can be used a powerful sensing device-free radio vision. Applications range from human body motion detection and localization to passive gesture recognition. In line with current evolution mobile phone [1], terminals not only ubiquitous...

Fall detection and localization of human operators inside a workspace are major issues in ensuring safe working environment. Recent research has shown that the perturbations radio-frequency (RF) signals commonly adopted for wireless communications can also be used as sensing tools device-free motion detection. Device-free RF-based applications range from tag-less body to monitoring well-being (e-Health). In this paper, we propose real-time system with special focus on joint fall The proposed...

Classical and centralized Artificial Intelligence (AI) methods require moving data from producers (sensors, machines) to energy hungry centers, raising environmental concerns due computational communication resource demands, while violating privacy. Emerging alternatives mitigate such high costs propose efficiently distribute, or federate, the learning tasks across devices, which are typically low-power. This paper proposes a novel framework for analysis of carbon footprints in distributed...

Low-complexity and privacy-respecting human sensing is a challenging task in smart environments as it requires the orchestration of multiple sensors, low-impact machine learning (ML) methods, resource-constrained Internet Things (IoT) devices. Client/server-based architectures are typically employed to support sensor fusion. However, these need data be moved to/from cloud or centers, which contrary fundamental requirement IoT applications limit costs, complexity, memory footprint,...

With the growing range of radio technologies and sensors used in next-generation industrial plants, radio-frequency sensing systems hold potential to generate high-value information. We discuss emerging tools, key technology enablers, distinct deployment experiences with a focus on communications beyond 5G future applications.

Safe human-machine interactions promote high flexibility in collaborative workspaces. Fall detection and localization of the operator are major issues ensuring a safe working environment. However, many proposed solutions not applicable for deployment industrial environments due to their performance limitations practical contexts. In this paper, we propose an integrated framework both fall operators inside shared workspace that employs radio-frequency (RF) signal analysis real-time. Multipath...

Electromagnetic (EM) body models predict the impact of human presence and motions on Radio-Frequency (RF) stray radiation received by wireless devices nearby. These may be co-located members a Wireless Local Area Network (WLAN) or even cellular connected with Wide (WAN). Despite their accuracy, EM are time-consuming methods which prevent adoption in strict real-time computational imaging problems Bayesian estimation, such as passive localization, RF tomography, holography. Physics-informed...

Subject counting systems are extensively used in ambient intelligence applications, such as smart home, building and retail scenarios. In this paper, we investigate the problem of transforming an unmodified WiFi radio infrastructure into a flexible sensing system for passive subject counting. We first introduce multi-dimensional channel features that capture presence. Then, compare Bayesian neural network based machine learning tools specialized discrimination Ensemble classification is to...

In cognitive radio networks (CRNs), primary-user (PUs) localization is a key aspect for improving network performance in terms of reliability and power adaptation. However, energy efficiency becomes critical issue CRNs due to the limited storage CR devices. this paper, we propose an energy-efficient cooperative algorithm (EE-CLA) PU positioning using mobile-aided CR. Since communication among CRs demanding, mobile manager introduced balance requirements positional accuracy consumption...

We propose a platform for the integration of passive radio sensing and vision technologies into cloud-IoT framework that performs real-time channel quality information (CQI) time series processing analytics. Radio allow to passively detect track objects or persons by using waves as probe signals encode 2D/3D view environment they propagate through. View reconstruction from received signals, CQI, is based on data tools, combine multiple measurements possibly heterogeneous IoT networks. The...

Industrial wireless networks are pushing towards distributed architectures moving beyond traditional server-client transactions. Paired with this trend, new synergies emerging among sensing, communications and Machine Learning (ML) co-design, where resources need to be across different field devices, acting as both data producers learners. Considering landscape, Federated (FL) solutions suitable for training a ML model in systems. In particular, decentralized FL policies target scenarios...

Device-free localization (DFL) systems have emerged in the last years as a powerful technology for tracking mobile targets without need of radio tags. Perturbations induced by moving objects on electromagnetic (EM) wavefield generated dense wireless network are measured and processed DFL system to track target trajectories. Despite several solutions been explored literature, mainly based fingerprinting approaches, deep understanding body-induced effects EM fields is still missing well...

Thermal vision systems based on low-cost IR array sensors are becoming attractive in many smart living scenarios. This paper proposes a Bayesian framework for recognition and discrimination of body motions real-time analysis thermal signatures. Unlike conventional frame-based methods, the proposed approach exploits statistical model extraction body-induced signatures mobility tracking multi-body inside an indoor area. prevents typical detection problems can be also used presence interfering...

Federated learning (FL) is emerging as a new paradigm for training machine model in cooperative networks. The parameters are optimized collectively by large populations of interconnected devices, acting learners that exchange local updates with the server, rather than user data. FL framework however centralized, it relies on server fusion and such limited single point failure. In this paper we propose distributed approach performs decentralized leveraging mutual cooperation between devices...

In cognitive radio networks (CRNs), knowledge of the primary users (PUs) position can be used to avoid harmful interference network, while at same time exploited improve CR performance. this paper, a localization algorithm is developed calculate PUs and novel location based (LCR) routing protocol proposed that has following properties: (i) it considers existence heterogeneous PUs, (ii) exploits information, (iii) jointly selects spectrum route, (iv) protects from interference. Clusters CRs...

In this paper, we show the possibility of using smartphone built-in cellular radio modem to track sudden changes in environment around it, thus turning cellphone into a radio-frequency (RF) virtual sensor. particular, demonstrate how isolate anomalous RF patterns by applying time series modeling and analysis downlink multi-cell signals. These anomalies may indicate situation change, namely, body or object(s), movement surrounding smart-phone. Unlike Wi-Fi Bluetooth devices, that can be...

Massive and unobtrusive screening of people in public environments is becoming a critical task to guarantee safety congested shared spaces, as well support early non-invasive diagnosis response disease outbreaks. Among various sensors Internet Things (IoT) technologies, thermal vision systems, based on low-cost infrared (IR) array sensors, allow track signatures induced by moving people. Unlike contact tracing applications that exploit short-range communications, IR-based sensing systems are...

In device-free radio frequency (RF) body occupancy inference systems, RF signals encode information (e.g., location, posture, activity) about moving targets (not instrumented) that alter the propagation in surroundings of link(s). Such systems are now getting more attention as they enable flexible location-based services for new smart scenarios spaces, safety and security, assisted living) just using off-the-shelf wireless devices. The goal this paper is to set fundamental signal processing...

Worker monitoring and protection in collaborative robot (cobots) industrial environments requires advanced sensing capabilities flexible solutions to monitor the movements of operator close proximity moving robots. Collaborative robotics is an active research area where Internet Things (IoT) novel technologies are expected play a critical role. Considering that no single technology can currently solve problem continuous worker monitoring, paper targets development IoT multisensor data fusion...

Recent advances in distributed learning raise environmental concerns due to the large energy needed train and move data to/from centers. Novel paradigms, such as federated (FL), are suitable for decentralized model training across devices or silos that simultaneously act both producers learners. Unlike centralized (CL) techniques, relying on big-data fusion analytics located hungry centers, FL scenarios collaboratively their models without sharing private data. This article breaks down...