This paper focuses on the realization of an Internet of Things (IoT) architecture to optimize waste management in the context of Smart Cities. In particular, a novel typology of sensor node based on the use of low cost and low power components is described. This node is provided with a single-chip microcontroller, a sensor able to measure the filling level of trash bins using ultrasounds and a data transmission module based on the LoRa LPWAN (Low Power Wide Area Network) technology. Together with the node, a minimal network architecture was designed, based on a LoRa gateway, with the purpose of testing the IoT node performances. Especially, the paper analyzes in detail the node architecture, focusing on the energy saving technologies and policies, with the purpose of extending the batteries lifetime by reducing power consumption, through hardware and software optimization. Tests on sensor and radio module effectiveness are also presented.

This paper discusses the architecture of a low cost Unmanned Surface Vehicle (USV) to be employed for the collection of crucial parameters about water quality in rivers, lakes or sea. The vehicle, called WeMo (Water Environmental Mobile Observer) has been realized exploiting off-the-shelf components and is provided with a modular array of sensors to measure chemical and physical parameters as well as to perform bathymetry. The low cost requirement is crucial since the vehicle is expected to be replicated in large quantities and then used for pervasive monitoring operations by providing it to local communities, administrations or even private stakeholders, in order to set up a sort of “Social Sensor Network”. In this sense, data analytics tools have also been introduced in order to automatically drive the vehicle along desired and suitable trajectories, and to process the collected data. These data can be used to estimate parameters of a mathematical model describing the ecological status of the monitored system. In particular, we apply an estimation procedure to a simple mathematical model of oxygen concentration in the water with explicit dependence on biophysical inputs. The estimation provides very satisfying performances, indeed the relative square error is less that 4∙10 -2. Moreover, once the vehicle is moving along a given trajectory, the status in the spatial domain can be reconstructed also in non monitored locations. The whole paper aims then at developing a complete monitoring ecosystem covering all the tasks of data collection, storage and analysis.

In this paper, a pervasive monitoring system to be deployed in underground environments is presented. The system has been specifically designed for the so-called “Bottini”, i.e., the medieval aqueducts dug beneath the Centre of Siena, Italy. The results of a measurement campaign carried out in the deployment scenario show that the transmission range of LoRa (Long Range) technology is limited to a maximum of 200 m, thus, making the adoption of a classical star topology impossible. Hence, a Linear Sensor Network topology is proposed based on multi-hop LoRa chain-type communications. In this scenario, an ad-hoc transmission scheme is presented that optimally evaluates the wake-up time of all nodes with the aim of minimizing the average energy dissipation deriving from clock offsets. Numerical results show that the proposed wake-up time optimization leads in the best case to a 50% reduction in power dissipation with respect to a scheme that evaluates the wake-up time in a non-optimal way.

Forest fires are the main cause of desertification, and they have a disastrous impact on agricultural and forest ecosystems. Modern fire detection and warning systems rely on several techniques: satellite monitoring, sensor networks, image processing, data fusion, etc. Recently, Artificial Intelligence (AI) algorithms have been applied to fire recognition systems, enhancing their efficiency and reliability. However, these devices usually need constant data transmission along with a proper amount of computing power, entailing high costs and energy consumption. This paper presents the prototype of a Video Surveillance Unit (VSU) for recognising and signalling the presence of forest fires by exploiting two embedded Machine Learning (ML) algorithms running on a low power device. The ML models take audio samples and images as their respective inputs, allowing for timely fire detection. The main result is that while the performances of the two models are comparable when they work independently, their joint usage according to the proposed methodology provides a higher accuracy, precision, recall and F${}_1$ score ($96.15$%, $92.30$%, $100.00$%, and $96.00$%, respectively). Eventually, each event is remotely signalled by making use of the Long Range Wide Area Network (LoRaWAN) protocol to ensure that the personnel in charge are able to operate promptly.

Sediments moved by wind, rivers, and waves round in the same way due to geometry.

This paper focuses on the comparison of transmission performances within critical environments (i.e., underwater, within metal enclosures and underground) for two of the most adopted enabling technologies for the Internet of Things (IoT): the Long Range Wide Area Network (LoRaWAN) protocol and the Narrowband Internet of Things (NB-IoT) standard. After a literature review, the two technologies are surveyed and compared. Then, the most exploited path loss models for the aforesaid application scenarios are presented. In order to assess and compare performances, a multi-protocol wireless sensor node prototype exploiting both the technologies for transmitting is described. Such prototype is then employed within field tests which consisted of underwater, through-metal and underground broadcasts with the aim of measuring losses only ascribed to the media. Eventually, tests results are analyzed and compared with the theoretical models.