Forests play a fundamental role in preserving the environment and fighting global warming. Unfortunately, they are continuously reduced by human interventions such as deforestation, fires, etc. This paper proposes and evaluates a framework for automatically detecting illegal tree-cutting activity in forests through audio event classification. We envisage ultra-low-power tiny devices, embedding edge-computing microcontrollers and long-range wireless communication to cover vast areas in the forest. To reduce the energy footprint and resource consumption for effective and pervasive detection of illegal tree cutting, an efficient and accurate audio classification solution based on convolutional neural networks is proposed, designed specifically for resource-constrained wireless edge devices. With respect to previous works, the proposed system allows for recognizing a wider range of threats related to deforestation through a distributed and pervasive edge-computing technique. Different pre-processing techniques have been evaluated, focusing on a trade-off between classification accuracy with respect to computational resources, memory, and energy footprint. Furthermore, experimental long-range communication tests have been conducted in real environments. Data obtained from the experimental results show that the proposed solution can detect and notify tree-cutting events for efficient and cost-effective forest monitoring through smart IoT, with an accuracy of 85%.
This paper deals with IoT devices deployed in remote areas without terrestrial Internet connectivity. We consider connecting IoT devices on the ground to the Internet through an aerial system based on an Unmanned Aerial Vehicle (UAV) for smart agriculture and environmental monitoring. The UAV flying over the remote area receives data from distributed IoT devices. The transmissions between the ground sensors and the UAV are carried out via LoRa. We have proposed a synchronization protocol for the opportunistic communication of LoRa IoT devices with a gateway onboard the UAV to save node battery life. Class A LoRa nodes on the ground transmit only when the UAV is expected to pass close to them; otherwise, they stay in the sleeping state most of the time. This paper provides a detailed description of the formulation of the synchronization protocol. The UAV’s flying dynamics have been considered for characterizing its speed and the time of visibility of each IoT sensor. Our model has allowed an analytical approach that can help to determine the best settings for LoRa transmissions. Finally, experiments have been carried out to assess the path loss attenuation, and a laboratory setup of the synchronization protocol has been implemented for the preliminary validation of our scheme.
18 August 2011 in International Journal of Wireless Information Networks
The Transmission Control Protocol (TCP) is one of the main communication protocols in the Internet, and it has been designed to provide an efficient reaction to packet loss events which are due to network congestion. Congestion is the main cause of losses in wired networks, but in today heterogeneous networks, loss events can also be introduced due to higher error rates on wireless channels, host mobility, and frequent handovers. Unfortunately, all packet losses are interpreted by TCP as a sign of congestion, triggering an inappropriate reaction which reduces its transmission rate and leads to performance degradation. In order to avoid this problem, it is important for TCP to correctly understand whether the reason of a packet loss is due to congestion or to a problem in the wireless link. This paper presents an innovative jitter-based cross-layer TCP algorithm, named XJTCP. It adopts the jitter ratio as loss predictor, joined with a layer two notification, in order to correctly infer the nature of a loss event. Performance evaluation and comparison with other common TCP implementations shows how XJTCP can be an interesting solution in the presence of wireless environments.
3 July 2020 in Universal Access in the Information Society
In this paper, the authors describe the realization and testing of a Wireless Sensor Network (WSN) framework aiming at measuring, remotely and in real time, the level variations of the sand layer of sandy beaches or dunes. The proposed framework is based on an innovative low cost sensing structure, able to measure the level variations with a 5-cm degree of precision and to locally transfer the acquired data through the ZigBee protocol. The described sensor is integrated in a wider ZigBee wireless sensor network architecture composed of an array of sensors that, arranged according to a grid layout, can acquire the same data at different points, allowing the definition of a dynamic map of the area under study. The WSN is connected to a local Global System for Mobile Communications (GSM) gateway that is in charge of data processing and transmission to a cloud infrastructure through a General Packet Radio Service (GPRS) connection. Data are then stored in a MySQL database and made available any time and anywhere through the Internet. The proposed architecture has been tested in a laboratory, to analyze data acquisition, processing timing and power consumption and then in situ to prove the effectiveness of the system. The described infrastructure is expected to be integrated in a wider IoT architecture including different typologies of sensors, in order to create a multi-purpose tool for the study of coastal erosive processes.