Low-Power Wide Area Networks (LPWANs), such as LoRaWAN, are pivotal for large-scale IoT deployments. However, traditional stationary gateways (GWs) impose scalability and cost constraints. We propose an AI-driven mobile GW architecture that leverages reinforcement learning (RL) to dynamically adapt GW mobility based on real-time network conditions. Our custom mobility module, integrated into OMNeT++ with FLoRa and TensorFlow Lite Micro, enables on-the-fly decision-making to optimize Packet Delivery Ratio (PDR) and fairness. The results of the simulations show that the introduced RL-GW achieves up to 99.94 % PDR in high-fidelity simulations, outperforming static and heuristic mobile strategies. The RL policy generalizes robustly across network sizes and scenarios, offering a scalable, low-overhead solution for adaptive LPWAN infrastructure.

Play a pivotal role in enabling Internet of Things (IoT) devices with sensing and actuation capabilities. Operating in remote and resource-constrained environments, these IoT devices face challenges related to energy consumption, crucial for network longevity. Existing clustering protocols often suffer from high control overhead, inefficient cluster formation, and poor adaptability to dynamic network conditions, leading to suboptimal data transmission and reduced network lifetime. This article introduces low-energy adaptive clustering hierarchy with reinforcement learning-based controller (LEACH-RLC), a novel clustering protocol designed to address these limitations by employing a mixed integer linear programming (MILP) approach for strategic selection of and node-to-cluster assignments. Additionally, it integrates a reinforcement learning (RL) agent to minimize control overhead by learning optimal timings for generating new clusters. LEACH-RLC aims to balance control overhead reduction without compromising overall network performance. Through extensive simulations, this article investigates the frequency and opportune moments for generating new clustering solutions. Results demonstrate the superior performance of LEACH-RLC over state-of-the-art protocols, showcasing enhanced network lifetime, reduced average energy consumption, and minimized control overhead. The proposed protocol contributes to advancing the efficiency and adaptability of WSNs, addressing critical challenges in IoT deployments.

The integration of artificial intelligence (AI) into embedded devices, a paradigm known as embedded artificial intelligence (eAI) or tiny machine learning (TinyML), is transforming industries by enabling intelligent data processing at the edge. However, the many tools available in this domain leave researchers and developers wondering which one is best suited to their needs. This paper provides a review of existing eAI tools, highlighting their features, trade-offs, and limitations. Additionally, we introduce EdgeMark, an open-source automation system designed to streamline the workflow for deploying and benchmarking machine learning (ML) models on embedded platforms. EdgeMark simplifies model generation, optimization, conversion, and deployment while promoting modularity, reproducibility, and scalability. Experimental benchmarking results showcase the performance of widely used eAI tools, including TensorFlow Lite Micro (TFLM), Edge Impulse, Ekkono, and Renesas eAI Translator, across a wide range of models, revealing insights into their relative strengths and weaknesses. The findings provide guidance for researchers and developers in selecting the most suitable tools for specific application requirements, while EdgeMark lowers the barriers to adoption of eAI technologies.

Time Slotted Channel Hopping (TSCH) is a medium access protocol defined in the IEEE 802.15.4 standard which have been proven to be one of the most reliable options when it comes to industrial applications. TSCH has been designed to be utilized in static network topologies. Thus, if an application scenario requires a mobile network topology, TSCH does not perform reliably. In this paper we introduce active connectivity for mobile application scenarios, such as mobile robots. This is a feature that enables the option to regulate physical characteristics such as the speed of a node as it moves, in order to keep being connected to the TSCH network. We model the active connectivity approach through a basic example where two nodes are moving towards the same direction to infer the main principles of the introduced approach. We evaluate the active connectivity feature through simulations and quantify trade-off between connectivity and application-layer performance.

TSCH is a MAC protocol able to provide reliable wireless communication to IoT networks. For applications where mobility is necessary research has shown that TSCH presents a suboptimal performance. One of the main issues is that the time required for a device to join a TSCH network is unpredictable and theoretically has no upper limit. The joining process to a TSCH network can take several minutes in some cases which can ensue significant delay and power overheads. In this paper we examine how TSCH can be adapted to support mobile applications, by improving the joining and re-joining procedure. Relevant information about the network topology is aggregated throughout the network to be utilized during the joining process. The results show that the joining time can be reduced by 50% and the re-joining time by 90%.

The Internet of Things (IoT) is evolving our society; however, the growing adoption of IoT devices in many scenarios brings security and privacy implications. Current security solutions are either unsuitable for every IoT scenario or provide only partial security. This paper presents AntibIoTic 2.0, a distributed security system that relies on Fog computing to secure IoT devices, including legacy ones. The system is composed of a backbone, made of core Fog nodes and Cloud server, a Fog node acting at the edge as the gateway of the IoT network, and a lightweight agent running on each IoT device. The proposed system offers fine-grained, host-level security coupled with network-level protection, while its distributed nature makes it scalable, versatile, lightweight, and easy to deploy, also for legacy IoT deployments. AntibIoTic 2.0 can also publish anonymized and aggregated data and statistics on the deployments it secures, to increase awareness and push cooperations in the area of IoT security. This manuscript recaps and largely expands previous works on AntibIoTic, providing an enhanced design of the system, an extended proof-of-concept that proves its feasibility and shows its operation, and an experimental evaluation that reports the low computational overhead it causes.

Distributed real-time systems often rely on time-triggered communication and task execution to guarantee end-to-end latency and time-predictable computation. Such systems require a reliable synchronized network time to be shared among end-systems. The IEEE 1588 Precision Time Protocol (PTP) enables such clock synchronization throughout an Ethernet-based network. While security was not addressed in previous versions of the IEEE 1588 standard, in its most recent iteration (IEEE 1588-2019), several security mechanisms and recommendations were included describing different measures that can be taken to improve system security and safety. One proposal to improve security and reliability is to add redundancy to the network through modifications in the topology. However, this recommendation omits implementation details and leaves the question open of how it affects synchronization quality. This work investigates the quality impact and security properties of redundant PTP deployment and proposes an observation window-based multi-domain, PTP end-system, design to increase fault-tolerance and security. We implement the proposed design inside a discrete-event network simulator and evaluate its clock synchronization quality using two test-case network topologies with simulated faults.

Over the past three decades large amounts of information have been converted to image formats from paper documents. Though in digital form, extracting the information, usually textual, from these documents requires complex image processing and optical character recognition techniques. The processing pipeline from the image to information typically includes an orientation correction task, document identification task, and text analysis task. When there are many document variants the tasks become difficult requiring complex sub-analysis for each variant and quickly exceeds human capability. In this work, we demonstrate a document analysis application with the orientation correction and document identification task carried out by supervised machine learning techniques for a large, international airline. The documents have been amassed over forty years with numerous variants and are mostly black and white, typically consist of text and lines, and some have extensive noise. Low level symbols are extracted from the raw images and separated into partitions. The partitions are used to generate statistical features which are then used to train the classifiers. We compare the classifiers for each task (e.g. decision tree, support vector machine, and random forest) to choose the most appropriate. We also perform feature selection to reduce the complexity of the document type classifiers. These parsimonious models result in comparable accuracy with 80% or fewer features.

Network monitoring has always being a challenge for network administrators. Their tasks are to keep track of every node in a network and ensure that they behave correctly. The Simple Network Management Protocol (SNMP) was designed to aid network administrators in their everyday tasks. A new wave of devices has been coming in the last years where small embedded devices are also being connected to the network. These new devices are part of the Internet of Things (IoT). IoT devices differ in many aspects from traditional networks. However, the main challenge with IoT devices is their hardware constraints (very limited memory and processing resources) and their limited energy budget. Indeed, they are usually powered by batteries and can be deployed in harsh environments. The current monitoring protocols were designed without taking the constraints of IoT devices into account. In an IoT context, there are very limited resources available for monitor constrained devices. While the current protocols do work with these devices they have an unnecessary overhead resulting in a waste of resources that could otherwise be used for some other task, or to save energy. This work proposes a new protocol designed for embedded devices and evaluates it against the state of the art.

The emerging Internet of Things has the potential to solve major societal challenges associated with healthcare provision. Low-power wireless protocols for residential Health Internet of Things applications are characterized by high reliability requirements, the need for energy-efficient operation, and the need to operate robustly in diverse environments in the presence of external interference. We enhance and experimentally evaluate the Time-Slotted Channel Hopping protocol from the IEEE 802.15.4 standard to address these challenges. Our contributions are a new schedule and an adaptive channel selection mechanism to increase the performance of time-slotted channel hopping in this domain. Evaluation in a test house shows that the enhanced system is suitable for our e-Health application and compares favorably with state-of-the-art options. The schedule provides higher reliability compared with the minimal scheduling function from the IETF 6TiSCH Working Group and has a better energy-efficiency/reliability tradeoff than the Orchestra scheduler. Results from 29 long-term residential deployments confirm the suitability for the application and show that the system is able to adapt and avoid channels used by WiFi. In these uncontrolled environments, the system achieves 99.96% average reliability for networks that generate 7.5 packets per second on average.

Cryptography is one of the most widely employed means to ensure confidentiality in the Internet of Things (IoT). Establishing cryptographically secure links between IoT devices requires the prior consensus to a secret encryption key. Yet, IoT devices are resource-constrained and cannot employ traditional key distribution schemes. As a result, there is a growing interest in generating secret random keys locally, using the shared randomness of the communicating channel. This article presents a secret key generation scheme, named SKYGlow, which is targeted at resource-constrained IoT platforms and tested on devices that employ IEEE 802.15.4 radios. We first examine the practical upper bounds of the number of secret bits that can be extracted from a message exchange. We contrast these upper bounds with the current state-of-the-art, and elaborate on the workings of the proposed scheme. SKYGlow applies the Discrete Cosine Transform (DCT) on channel observations of exchanged messages to reduce mismatches and increase correlation between the generated secret bits. We validate the performance of SKYGlow in both indoor and outdoor scenarios, at 2.4 GHz and 868 MHz respectively. The results suggest that SKYGlow can create secret 128-bit keys of 0.9978 bits entropy with just 65 packet exchanges, outperforming the state-of-the-art in terms of energy efficiency.

We are interested in hot and cold water flow detection in domestic kitchen and bathroom taps for smart home environments. Water flow monitoring is particularly valuable for long-term behavioural monitoring systems for health-related applications, as it enables the collection of long-term data on the hydration levels of the house residents, and it is associated with several activities of daily life, such as cooking and cleaning. This paper presents a water flow sensing device that is based on sensing the vibrations on the pipe when water is flowing through them. The proposed solution is noninvasive and energy efficient, as it does not require cutting the water pipes or altering the plumbing system, and consumes less then 2 uA in continuous operation. The proposed water flow sensor has been integrated to SPHERE, a sensing platform of non-medical sensors for healthcare monitoring and behavioural analytics in a home environment, and deployed to more than 15 residential properties.

Networks deployed in real-world conditions have to cope with dynamic, unpredictable environmental temperature changes. These changes affect the clock rate on network nodes, and can cause faster clock de-synchronization compared to situations where devices are operating under stable temperature conditions. Wireless network protocols, such as time-slotted channel hopping (TSCH) from the IEEE 802.15.4-2015 standard, are affected by this problem, since they require tight clock synchronization among all nodes for the network to remain operational. This paper proposes a method for autonomously compensating temperature-dependent clock rate changes. After a calibration stage, nodes continuously perform temperature measurements to compensate for clock drifts at runtime. The method is implemented on low-power Internet of Things (IoT) nodes and evaluated through experiments in a temperature chamber, indoor and outdoor environments, as well as with numerical simulations. The results show that applying the method reduces the maximum synchronization error more than ten times. In this way, the method allows reduction in the total energy spent for time synchronization, which is practically relevant concern for low data rate, low energy budget TSCH networks, especially those exposed to environments with changing temperature.

Wireless industrial networks require reliable and deterministic communication. Determinism implies that there must be a guarantee that each data packet will be delivered within a bounded delay. Moreover, it must ensure that the potential congestion or interference will not impact the predictable properties of the network. In 2016, IEEE 802.15.4 time-slotted channel hopping (TSCH) emerged as an alternative medium access control to the industrial standards such as WirelessHART and ISA100.11a. However, TSCH is based on traditional collision detection and retransmission, and cannot guarantee reliable delivery within a given time. This paper proposes LeapFrog Collaboration (LFC) to provide deterministic and reliable communication over a routing protocol (RPL) based network. LFC is a novel multipath routing algorithm that takes advantage of route diversity by duplicating the data flow onto an alternate path. Simulations and analytical results demonstrate that LFC significantly outperforms the single-path retransmission-based approach of RPL + TSCH and the state-of-the-art LinkPeek solution.

The upcoming Internet of Things (IoT) applications include real-time human activity monitoring with wearable sensors. Compared to the traditional environmental sensing with low-power wireless nodes, these new applications generate a constant stream of a much higher rate. Nevertheless, the wearable devices remain battery powered and therefore restricted to low-power wireless standards such as IEEE 802.15.4 or Bluetooth Low Energy (BLE). Our work tackles the problem of building a reliable autonomous schedule for forwarding this kind of dynamic data in IEEE 802.15.4 TSCH networks. Due to the a priori unpredictability of these data source locations, the quality of the wireless links, and the routing topology of the forwarding network, it is wasteful to reserve the number of slots required for the worst-case scenario, under conditions of high expected datarate, it is downright impossible. The solution we propose is a hybrid approach where dedicated TSCH cells and shared TSCH slots coexist in the same schedule. We show that under realistic assumptions of wireless link diversity, adding shared slots to a TSCH schedule increases the overall packet delivery rate and the fairness of the system.

The confidentiality of communications in the Internet of Things (IoT) is critical, with cryptography currently being the most widely employed method of ensuring it. Establishing cryptographically secure communication links between two transceivers requires the pre-agreement on some key, unknown to an external attacker. In recent years there has been growing attention in techniques that generate a shared random key through observation of the channel and its effects on the exchanged messages. In this work we present SKYGlow, a novel scheme for secret-key generation, designed for IoT devices, such as IEEE 802.15.4 and Bluetooth Low Energy (BLE) transceivers. SKYGlow employs the Discreet Cosine Transform (DCT) of channel observations and Slepian-Wolf coding for information reconciliation. Real-life experiments have resulted in the creation of 128-bit secret keys with only 65 packet exchanges and with an entropy of 0.9978 bits, making our scheme much more energy-efficient compared with others in the existing literature.

Advance clustering techniques have been widely used in Wireless Sensor Networks (WSNs) since they can potentially reduce latency, improve scheduling, decrease end-to-end delay and optimise energy consumption within a dense network topology. In this paper, we present a novel clustering algorithm for high density IEEE 802.15.4-2015 Time-Slotted Channel Hopping (TSCH). In particular, the proposed methodology merges a variety of solutions into an integrated clustering design. Assuming an homogeneous network distribution, the proposed configuration deploys a hierarchical down-top approach of equally numbered sub-groups, in which the formation of the separate sub-groups is adapted to the network density and the node selection metric is based on the link quality indicator. The presented algorithm is implemented in Contiki Operating System (OS) and several test vectors have been designed in order to evaluate the performance of the proposed algorithm in a COOJA simulation environment. Performance results demonstrate the capability of the clustering structure since compared to the default scheme it significantly improves the energy efficiency up to 35%, packet drops more than 40% as well the packet retransmission rate. Last but not least, the outcome of this study indicates a major increase in the network lifetime, i.e., up to 50%.

With the ubiquity of sensing technologies in our personal spaces, the protection of our privacy and the confidentiality of sensitive data becomes a major concern. In this letter, we focus on wearable embedded systems that communicate data periodically over the wireless medium. In this context, we demonstrate that private information about the physical activity levels of the wearer can leak to an eavesdropper through the physical layer. Indeed, we show that the physical activity levels strongly correlate with changes in the wireless channel that can be captured by measuring the signal strength of the eavesdropped frames. We practically validate this correlation in several scenarios in a real residential environment, using data collected by our prototype wearable accelerometer-based sensor. Finally, we propose a privacy enhancement algorithm that mitigates the leakage of this private information.

In this paper, error correction is introduced to the Bluetooth low energy and IEEE 802.15.4 standards by utilizing data redundancy provided by cyclic redundancy check (CRC) codes used by both protocols to detect erroneous packets. A scenario with an energy-constrained transmitter and a constraint-free infrastructure is assumed that enables additional signal processing at the receiving side, keeping the transmitter intact. CRC error correction is achieved using a novel approach of applying iterative decoding techniques. The proposed methods are evaluated based both on simulated and real packets. It is shown that by enabling CRC error correction, up to 2.5 dB of the signal to noise ratio gain can be achieved, while up to 35% of real corrupted packets can be corrected, at no extra cost for the transmitter. This results in potential range extension and longer battery life caused by a reduced number of retransmissions.

Wearable technologies play a key role in the shift of traditional healthcare services towards eHealth and self-monitoring. Maintenance overheads, such as regular battery recharging, impose a limitation on the applicability of such technologies in some groups of the population. In this paper, we propose an activity monitoring system that is based on wearable sensors that are powered by textile inductive charging surfaces. By strategically positioning these surfaces on pieces of furniture that are routinely used, the system passively charges the wearable sensor whilst the user is present. As a proof-of-concept example, experiments conducted on a prototype implementation of the system suggest that 36 minutes of daily desktop computer usage are on average sufficient to maintain a wearable sensor energy neutral.

The IEEE 802.15.4-2015 standard defines a number of Medium Access Control (MAC) layer protocols for low-power wireless communications in the IoT. Originally defined in the IEEE 802.15.4e amendment, TSCH (Time Slotted Channel Hopping) is among the proposed mechanisms. TSCH is a scheme aiming to guarantee network reliability by keeping nodes time-synchronised at the MAC layer. In order to ensure successful communication between a sender and a receiver, the latter starts listening shortly before the expected time of a MAC layer frame's arrival. The offset between the time a node starts listening and the estimated time of frame arrival is called guard time and it aims to reduce the probability of missed frames due to clock drift. In this poster, we investigate the effect of the guard time duration on energy consumption. We identify that, when using the 6tisch minimal schedule, the most significant cause of energy consumption is idle listening during guard time. Therefore, the energy-efficiency of TSCH can be significantly improved by guard time optimisation. Our performance evaluation results, conducted using the Contiki operating system, show that an efficient configuration of guard time may reduce energy consumption by up to 30%, without compromising network reliability.

This paper presents 3D radiation pattern analyses for omnidirectional (dipole) and directional (patch) antennas for various body locations encompassing wrist orientations and chest positions when mounted on a body phantom. In addition to analysing the directivities and relative efficiencies at 2.44GHz for different body positions, the study considers Sector and Slice analysis of the radiation patterns. In Sector analysis, the directivity is averaged for 12 azimuth-elevation sectors, while in Slice analysis, it is averaged for 28 azimuth sectors over the full elevation. It is shown that the antenna efficiency due to body blockage can be as low as 23% relative to the chest position efficiency, and directivities ranges from 5.4 to 10.5dBi for the antennas at different orientations. The Sector analysis identifies highest average signal levels, which are between table and door height for the dipole and above door height for the patch. The Slice analysis, which doesn't account for access point or user heights, shows average directivities that peak at 5.1 and 4.4dBi for the dipole and patch antenna, respectively. Using these antennas as part of a switch diversity system would improve the average directivity by approximately 7dBi in regions of low signal levels, and hence are potentially useful for wristbands and smart clothing.

In this paper, we investigate the application of iterative decoding algorithms to error correction of cyclic redundancy check (CRC) codes widely used in low-energy communication standards. We consider the case when traditional error correction codes are not available due to energy constraints at the transmitter. Using the CRC-24 code adopted by the Bluetooth Low Energy standard as an example, we show how two iterative techniques traditionally used for decoding of low-density parity check codes - Belief Propagation (BP) and the Alternating Direction Method of Multipliers (ADMM) - can be applied to the high-density parity check matrix of the code. The performance of both techniques is evaluated through simulation, and it is demonstrated that a gain of up to 1.7 dB in terms of the SNR per bit and a total reduction of the packet error rate by more than 70% can be achieved compared with the non-correction scenario, at no extra cost for the transmitter. We also compare the two techniques and use the standard syndrome look-up method as a benchmark. Both schemes enable the correction of multiple errors, with the ADMM-based decoder demonstrating better overall performance than BP.

We consider an environment where self-interested IEEE 802.11 Wireless Local Area Networks (WLANs) have overlapping coverage, and investigate the incentives that can trigger handovers between the WLANs. Our focus is on the incentives for supporting handovers due solely to the improved performance for all wireless networks. Such incentives arise due to a well-known property of 802.11 networks, where low rate users that send traffic significantly degrade the performance of high rate users that are associated with the same access point. A key difference of this paper compared to other works is that WLANs are self-interested, seeking to improve the performance of their own clients. We develop a comprehensive analytical model for accurately identifying and quantifying the handover gains. The model captures cases such as uplink and downlink traffic, wired link capacity constraints, and nonsaturated traffic conditions, and yields a practical handover decision policy. Simulations and experiments on a real testbed verify the accuracy of the model, and indicate that significant gains can be achieved through performance-induced handover incentives.

In this work we present "AlertMe", a semantics-based, context-aware notification system that provides personalized alerts to graduate students based on their preferences. An extensive description of the system is carried out. We present the underlying ontology that models the available knowledge, as well as how higher level knowledge inference and context-based decision making is achieved through rule-based reasoning. Finally, we outline the technical aspects of the developed system, covering issues involving the integration of the various subcomponents.