Daniele Trinchero

Recently, mycotoxin prediction has mainly relied on meteorological data and crop physiology. The contribution of soil characteristics as additional environmental variables remains largely unexplored. A systematic literature search was carried out to analyze the latest research (from 2020 to 2025) on the relationship between soil properties (temperature, water content, pH, and electrical conductivity), fungal communities (particularly Aspergillus and Fusarium), and different crops (mainly peanut, wheat, and maize). Measurement methodologies were analyzed, with a focus on the use of in-field soil sensors in correlation studies and predictive models. Disease incidence and mycotoxin occurrence were related to stressful soil conditions, such as different pH levels, wetness or drought, and temperatures above 25 °C. Other external variables (crop and field management) must also be considered. Laboratory equipment was primarily used in correlation studies, with limited in-field sensor implementation. Although recent predictive models included soil properties as effective inputs, they mostly relied on satellite data. However, real-time conditions and fluctuations, which can be captured by in-field soil sensors, are essential for training new functional models. To monitor soil properties, IoT technologies must be considered, but their implementation is still not sufficient to collect widespread data. Therefore, groundwork is needed to fill this gap with high-quality soil data for future in-field experimentation.

Many sciences exploit algorithms in a large variety of applications. In agronomy, large amounts of agricultural data are handled by adopting procedures for optimization, clustering, or automatic learning. In this particular field, the number of scientific papers has significantly increased in recent years, triggered by scientists using artificial intelligence, comprising deep learning and machine learning methods or bots, to process field, crop, plant, or leaf images. Moreover, many other examples can be found, with different algorithms applied to plant diseases and phenology. This paper reviews the publications which have appeared in the past three years, analyzing the algorithms used and classifying the agronomic aims and the crops to which the methods are applied. Starting from a broad selection of 6060 papers, we subsequently refined the search, reducing the number to 358 research articles and 30 comprehensive reviews. By summarizing the advantages of applying algorithms to agronomic analyses, we propose a guide to farming practitioners, agronomists, researchers, and policymakers regarding best practices, challenges, and visions to counteract the effects of climate change, promoting a transition towards more sustainable, productive, and cost-effective farming and encouraging the introduction of smart technologies.

The paper presents a double-radio wireless multimedia sensor node (WMSN) with a camera on board, designed for plant proximal monitoring. Camera sensor nodes represent an effective solution to monitor the crop at the leaf or fruit scale, with details that cannot be retrieved with the same precision through satellites or unnamed aerial vehicles (UAVs). From the technological point of view, WMSNs are characterized by very different requirements, compared to standard wireless sensor nodes; in particular, the network data rate results in higher energy consumption and incompatibility with the usage of battery-powered devices. Avoiding energy harvesters allows for device miniaturization and, consequently, application flexibility, even for small plants. To do this, the proposed node has been implemented with two radios, with different roles. A GPRS modem has been exclusively implemented for image transmission, while all other tasks, including node monitoring and camera control, are performed by a LoRaWAN class A end-node that connects every 10 min. Via the LoRaWAN downlink, it is possible to efficiently control the camera settings; the shooting times and periodicity, according to weather conditions; the eventual farming operations; the crop growth stages and the season. The node energy consumption has been verified in the laboratory and in the field, showing that it is possible to acquire one picture per day for more than eight months without any energy harvester, opening up further possible implementations for disease detection and production optimization.

An innovative implementation of an electronic leaf wetness sensor (LWS) is proposed. It utilizes capacitive sensing, combined with an innovative data acquisition method, which implements a capacitance-to-digital converter. The study explores the design procedure of a capacitive LWS, proposing an analytical approach and emphasizing low manufacturing costs. Since the LWS is intended for Internet-of-Things applications, this article estimates its energy consumption, introducing a boost regulator to optimize power usage, contributing to extend the battery life. The study presents simulation results and experimental validations, including an ad-hoc calibration procedure under controlled conditions. The sensors were tested in real agricultural environments over a complete vegetative season, demonstrating their capability to operate continuously without problems.

Precision agriculture, driven by the convergence of smart sensors and advanced technologies, has emerged as a transformative force in modern farming practices. The present review synthesizes insights from a multitude of research papers, exploring the dynamic landscape of precision agriculture. The main focus is on the integration of smart sensors, coupled with technologies such as the Internet of Things (IoT), big data analytics, and Artificial Intelligence (AI). This analysis is set in the context of optimizing crop management, using resources wisely, and promoting sustainability in the agricultural sector. This review aims to provide an in-depth understanding of emerging trends and key developments in the field of precision agriculture. By highlighting the benefits of integrating smart sensors and innovative technologies, it aspires to enlighten farming practitioners, researchers, and policymakers on best practices, current challenges, and prospects. It aims to foster a transition towards more sustainable, efficient, and intelligent farming practices while encouraging the continued adoption and adaptation of new technologies.

We explore THz communication uplink multi-access with multi-hop Intelligent reflecting surfaces (IRSs) under correlated channels. Our aims are twofold: 1) enhancing the data rate of a desired user while dealing with interference from another user and 2) maximizing the combined data rate. Both tasks involve non-convex optimization challenges. For the first aim, we devise a sub-optimal analytical approach that focuses on maximizing the desired user’s received power, leading to an over-determined system. We also attempt to use approximate solutions utilizing pseudo-inverse (Pinv) and block solution (BLS) based methods. For the second aim, we establish a loose upper bound and employ an exhaustive search (ES). We employ deep reinforcement learning (DRL) to address both aims, demonstrating its effectiveness in complex scenarios. DRL outperforms mathematical approaches for the first aim, with the performance improvement of DDPG over the block solution ranging from 8% to 57.12%, and over the pseudo-inverse ranging from 41% to 190% for a correlation-factor equal to 1. Moreover, DRL closely approximates the ES for the second aim. Furthermore, our findings show that as channel correlation increases, DRL’s performance improves, capitalizing on the correlation for enhanced statistical learning.

The majority of fungal infections are caused by leaf wetness, due to excessive humidity, mist, rain or irrigation. Leaf wetness duration (LWD), the total number of hours per day in which water is present on the foliage, can be monitored through a leaf wetness sensor (LWS). In this paper, an innovative procedure to manufacture a precise, low power and low cost electronic sensor is proposed to provide a tool for agronomic exploitation. The measurement subsystem consists of a two-side capacitive leaf wetness sensor, where a water presence varies the dielectric constant of the surrounding medium. The data acquisition subsystem exploits a capacitance-to-digital converter. Experimental results in controlled and real environments confirm reliability.

THz technology is considered a key element in 6G wireless communication because it provides ultra-high bandwidths, considerable capacities, and significant gains. However, wireless systems operating at high frequencies are faced with uncertainty and highly dynamic channels. Reflecting intelligent surfaces (RISs) can increase the range of the THz communication links and boost the rate at the receiver. In contrast to the existing literature, we investigate the scenario of multiple access multi-hop (cascaded) RISs uplink THz networks in a correlated channel environment. We show that our inspected cascaded RIS system is over-determined and that the rate maximization optimization problem is non-convex. To this end, we derive a closed-form expression of the received power and derive an analytical solution based on pseudo-inverse to obtain optimum RISs’ phase shifts that maximize the received signal power and hence increase the rate. In addition, we utilize deep reinforcement learning (DRL), which is capable of solving non-convex optimization problems, to obtain the optimum cascaded RISs’ phase shifts at the receiver taking into account the situation of the spatially correlated channels. Simulation results demonstrate that the DRL algorithm achieves higher rates than the mathematical sub-optimal method and the case of randomized phases.