Cintia Borges Margi

Soil moisture monitoring is essential for efficient irrigation and crop management in precision agriculture. Conventional methods rely on manual sampling or dedicated sensors, which limit large-scale deployment due to cost and infrastructure requirements. RF-based sensing offers an alternative by exploiting changes in signal amplitude and phase caused by soil dielectric properties. This work evaluates soil moisture sensing using Wi-Fi on resource-constrained ESP32 devices in a laboratory and an in the field. Our study takes into consideration the distance between devices, and the CSI amplitude and phase. Confined experiments introduced bias in channel metrics, while field tests revealed sensing limitations with increasing range.

Rural roads are vital for the transportation of persons and goods, but are vulnerable to animal-vehicle collisions, which cause substantial environmental, economic, and public health impacts every year. Wi-Fi sensing has recently emerged as a cost-effective approach to detecting hazardous situations on roads, overcoming some limitations of traditional monitoring systems. However, scaling Wi-Fi sensing to cover long road sections introduces challenges, including interference between Wi-Fi nodes and with other protocols operating in the same industrial, scientific, and medical (ISM) bands, such as IEEE 802.15.4. In this work, we present a scalable, low-cost architecture for Wi-Fi-based monitoring of rural roads. We develop and optimize machine learning models for real-time detection and classification of pedestrians, vehicles, and animals using low-cost Internet of Things (IoT) devices, achieving over 95% accuracy and a miss rate of only 0.7%. We evaluate the models under varied conditions, including different sensor distances, light rain, and previously unseen rural locations, demonstrating robustness across environmental variations. In addition, we conduct network simulations to assess the coexistence of Wi-Fi sensing with IEEE 802.15.4 communications and perform LoRaWAN experiments in rural environments to validate long-range data transmission. We also release an open dataset containing all evaluated scenarios to support reproducibility and further research. Our results confirm the feasibility of the proposed system in terms of sensing accuracy, network performance, and deployment cost.

Road traffic accidents involving animals cause great health, environmental, and monetary costs every year, specially on rural areas. Current animal detection systems suffer from either cost, scalability or accuracy issues, which prevent their effective use in a more extensive manner. In our research, we explore WiFi sensing to monitor events in rural roads, using low-cost IoT devices to collect WiFi data, publishing the first open CSI dataset of animal crossings and applying machine learning techniques to detect and classify them. Event detection runs on ESP32-S3 devices, while classification runs on Raspeberry Pi 4 devices, both with accuracy of at least 95%. Our system enables a scalable and cost-effective solution for monitoring multiple kilometers of roads.

LoRaWAN is a widely adopted protocol for Internet of Things applications with energy constraints and large coverage area demands. The protocol standardization defines a star-of-stars topology, where end-devices send their packets directly to a gateway that forwards the messages to the network server, where they are handled depending on the application. LoRaWAN is at the top of the Long Range (LoRa) physical layer based on the Chirp Spread Spectrum (CSS) modulation, where signals are spread along the frequency by the Spreading Factor (SF). The modulation technique gives the LoRaWAN long range coverage characteristics, that could reach 45 kilometers in theory. Furthermore, the protocol adopts the Industrial, Medical, and Scientific (ISM) frequency band plans, which impose spectrum occupation limitations, and the constraints are defined by a device Duty Cycle (DC). The DC imposes a restriction of 0.1% up to 10% in transmission time, limiting the total amount of messages an end-deceive could send. In order to increase even more the coverage area, many works proposed the adoption of multihop techniques in LoRaWAN networks. Proposed solutions are based on routing and relay approaches, both with benefits and limitations. However, the literature does not explore the DC constraints on the multihop LoRaWAN solutions since packets might be enqueued. The packet delay increases due to the time spent in the buffer, and the delivery ratio decreases because packets are dropped when the buffer is full. To understand such impact, we provide a model to generalize the DC behavior on multihop LoRaWAN networks, focused on the packet delivery ratio and delay metrics. We evaluate our model analytically and experimentally. First, we conducted a numerical comparison between the standard network with the multihop Long Range Wide Area Network (LoRaWAN) over duty cycle constraints. The comparison considers a rural area and the coverage area limited by the 802.11 ah propagation model. The results elucidated the tradeoff between packet size, number of packets, and delay. The two-hop solution could decrease the delay compared with direct transmission using high spreading factors. However, in dense networks, the intermediate device buffer increases the packet delivery delay. It could reduce the packet delivery ratio because the device must respect the duty cycle to forward packets. On the other hand, higher spreading factors decrease the payload size and need more packets to transmit the same message. Then, we compared analytical and experimental results in four relayed scenarios with different numbers of devices and different intermediate device buffer sizes. The results show the DC significantly increases the network delay and packet losses. The intermediate device also causes packet prioritization, making the network unfair. The multihop strategies are relevant for many applications, including coverage area extension, but the solutions must consider the DC constraints as a limitation to their adoption.

Metaverse is a virtual universe connecting virtual worlds and entities to the real world. The Metaverse applications are based on real world data to provide immersion and integration to users. IoT is a network interconnection paradigm designed for heterogeneous data collection and devices with common objectives. The perspective to integrate both technologies seems logical by the presented definition. However, there are challenges to overcome, such as an architecture to describe the elements and how they work together. Our work presents the architecture solutions on Metaverse and IoT and then presents a three-layer architecture proposal with IoT as a Metaverse service. The article also provides an overview of network, time, and standardization challenges. We conclude the Metaverse-IoT confluence is an opportunity for research and development, besides the technology’s opposite requirements.

Road traffic accidents involving animals cause great health, environmental and monetary costs every year, specially on rural areas. Current animal detection systems suffer from either cost, scalability or accuracy issues, which prevent their effective use in a more extensive manner. To provide an accurate and cost-effective detection system, we employ WiFi sensing using low-cost IoT devices to train a modern deep learning model - the Transformer network. Our Transformer network detects road crossings with an accuracy of 97.1 percent and exhibits low false positive and false negative rates. In particular, our model accurately distinguishes vehicles from small and large animals, which enables scalable and economical accident prevention over large distances in rural roads. Finally, our model is on par with state-of-the-art predictive models, outperforming recent AutoML mechanisms and evidencing the suitability of the Transformer network for WiFi sensing-based tasks.

Given the improvements to network flexibility and programmability, software-defined wireless sensor networks (SDWSNs) have been paired with IEEE 802.15.4e time-slotted channel hopping (TSCH) to increase network efficiency through slicing. Nonetheless, ensuring the quality of service (QoS) level in a scalable SDWSN remains a significant difficulty. To solve this issue, we introduce the application-aware (AA) scheduling approach, which isolates different traffic types and adapts to QoS requirements dynamically. To the best of our knowledge, this approach is the first to support network scalability using shared timeslots without the use of additional hardware while maintaining the application’s QoS level. The AA approach is deeply evaluated compared with both the application traffic isolation (ATI) approach and the application’s QoS requirements using the IT-SDN framework and by varying the number of nodes up to 225. The evaluation process took into account up to four applications with varying QoS requirements in terms of delivery rate and delay. In comparison with the ATI approach, the proposed approach enhanced the delivery rate by up to 28% and decreased the delay by up to 57%. Furthermore, even with four applications running concurrently, the AA approach proved capable of meeting a 92% delivery rate requirement for up to 225 nodes and a 900 ms delay requirement for up to 144 nodes.

Software-Defined Networking has been used to leverage security solutions for wireless sensor networks. However, this paradigm turns networks vulnerable to distributed denial of service attacks. IDIT-SDN is a tool for Software-defined Wireless Sensor Networks devised for DoS and DDoS attacks simulation and detection. This tool provides a framework for anomaly detection and a communication protocol to share security wise information from the sensor network to the controller. We demonstrate its use by showing a cooperative DDoS attack detection and attacker identification application based on distributed (every node) and centralized (controller) anomaly detection.

LoRaWAN is one of the protocols responsible for connecting devices to the IoT ecosystem. The protocol provides connectivity with low energy consumption and a large coverage area. However, many scenarios impose difficulties on the transmission, and sometimes it is necessary to increase the coverage using a multihop approach. The objective of this work is to introduce a gateway-based relay device for a multihop LoRaWAN network. The called relay-gateway is an intermediate device capable of receiving data from many sources due to its hardware capabilities. It is reliable in terms of energy consumption because it uses a fixed energy source. The experimental evaluation conducted at the university campus shows that the relay solution is suitable for extending the network and providing connectivity to remote areas. The results show a 50% increase of packet delivery for some places using the relay solution compared with direct transmission using SF7.