Search

Proximity detection is the process of estimating the closeness between a target and a point of interest, and it can be estimated with different technologies and techniques. In this paper we focus on how detecting proximity between people with a TinyML-based approach. We analyze RSS values (Received Signal Strength) estimated by a micro-controller and propagated by Bluetooth's tags. To this purpose, we collect a dataset of Bluetooth RSS signals by considering different postures of the involved people. The dataset is adopted to train and test two neural networks: a fully-connected and an LSTM model that we compress to be executed directly on-board of the micro-controller. Experimental results conducted over the dataset show an average precision and recall metrics of 0.8 with both of the models, and with an inference time less than 1 ms.

Final technical report from the ITU-T Focus Group on Environmental Efficiency for Artificial Intelligence and other Emerging Technologies (FG-AI4EE), discussing the need for integrating and harmonizing environmental, social models and sustainability needs when designing AI-IoT based solutions. The report highlights (I) current barriers hampering the adoption of a comprehensive path that addresses all three needs, (II) the risks stemming from single-path sustainability approaches, and (III) provides suggestions for future work that can foster and promote the adoption of a more comprehensive process of designing sustainable AI-IoT systems.

Mobile CrowdSensing (MCS) is a computational paradigm designed to gather sensing data by using the personal devices of the MCS platform users. However, being the mobility of the devices tightly correlated with mobility of their owners, the covered area might be limited to specific sub-regions. We extend the coverage capability of a MCS platform by exploiting unmanned aerial vehicles (UAV) as mobile sensors gathering data from low covered locations. We present a probabilistic model designed to measure the coverage of a location by analysing the user's trajectories and the detouring capability of MCS users towards a location of interest. Our model provides a coverage used revealing low-covered locations. These are used as targets for StationPositioning, our proposed algorithm optimizing the deployment of k UAV stations. We analyze the performance of StationPositioning by comparing the ratio of the covered locations against Random, DBSCAN and KMeasn algorithm. We explore the performance by varying the time period, the deployment regions and the existence of areas where it is not possible to deploy any station. Our experimental results show that StationPositioning is able to optimize the selected target location for a number of UAV stations with a maximum covered ratio up to 60%

Fertigation represents one of the most efficient techniques to optimize the delivery of nutrients and water to the crop. It may preserve the cultivation system from nutrient leaching that is harmful for the environment. However, an empirical estimation of the actual crop requirements, in intensive vegetable production systems, may lead to a large waste of nutrients and water even if microirrigation is adopted thus making useless the adoption of such a technology. Vegetable crops are very sensitive to fertigation and an incorrect management can easily limit growers' incomings since defects or excesses are both detrimental for the achievement of high yield and quality. Leafy vegetables, for example, may accumulate excess nitrate in the harvested product, which are armful for human health, or show poor shelf life in presence of uneven distribution of fertilisers and water throughout the cultivation period. All the above aspects pose a threat for the economic and environmental sustainability of leafy vegetables. Many different tools for precision agriculture nowadays aim at improving nutrient and water use efficiency in intensive cropping systems. Simulation models may offer some characteristics that make growers' decisions easier and faster. These tools can be integrated in decision support systems to assist growers in combination with sensing technologies for monitoring plant nutrient and water status. In this work we calibrated a photosynthesis-based model to simulate plant growth thus providing fundamental parameters for the estimation of crop evapotranspiration and nutrient uptake such as leaf area index and biomass accumulation. The final output of the model is a dynamic simulation of the macronutrient concentrations to be maintained by the fertigation system to match the actual plant needs. Experimental data were collected on lettuce (Lactuca sativa var. capitata L.; Salanova®-type) grown on rockwool slabs, under optimal (i.e., not limiting) growing conditions, with a plant density of 22 pt/m2 . The calibration was based on different cultures carried out during the 2018-2019 autumn-winter-spring season under greenhouse conditions. The analysis of simulated versus measured data showed a high capability of the model to predict crop requirements.

In order to explore short-term facilitation of the Schaffer collateral to CA1 synapse in mouse hippocampal brain slices, we measured the time course of the decay of the peak amplitude of successive EPSCs during progressive MK-801-dependent block (PMDB) of NMDAR responses to paired (R1 and R2) stimuli. We made the unexpected observation that the R2 response exhibited a slower PMDB decay constant than that of the R1 response. This indicated that the facilitated R2 response engages release sites with NMDARs that are protected from opening and consequent MK-801 block during the basal R1 response. We then utilized conditions that affect synaptic glutamate distribution to dissect the components of the distinct PMDB decay constants of the first and second of paired pulses. While extra-synaptic NMDARs and glutamate transporters appear to play only minor roles in the differences of the PMDB decay constant, we showed important roles for the R1 response itself and for glutamate diffusion in determining the PMDB decay constant of R2. We used a simple computational model with realistic parameters that allowed us to predict the time course of R2 decay based on the R1 decay time course.

Locomotion system issues are major issues in dairy herds, affecting both animal welfare and farm productivity. Early detection could improve the effectiveness of treatments and increase the chances to cure lame cows. Currently, locomo- tion issues detection requires direct observation of cows walk- ing (locomotion score). However, this is a time-consuming task and is not always an available option in large dairy farms. Aim of this preliminary study was to build a predictive model for locomotion system issues in Italian Holstein dairy cows using some novel phenotypes from automatic recording systems (milking parallel parlour and SCR Heatime and DataFlow2 system) as predictors. Data was recorded from a commercial farm located in the province of Mantua (Lombardy, Northern Italy) for a total of 413 animals, daily monitored for two years (Sept. 2014 - Dec. 2016). The response variable was binary (0/1: healthy and diseased, respectively). The selected variables were daily rumination time, parity, DIM, daily milk production, daily activity and month of recording. Summary statistics (mean and SD) were calculated: rumination time, 563.28 ± 88.48 min/day; parity, 1.93 ± 1.28; DIM, 171.34 ± 124.14 days; milk production, 24.14 ± 13.74 kg/day; activity, 614.00 ± 134.64 min/day. DIM were classified in four classes to assess the potential effect of the lactation stage: dry, early, mid, and late lactation (no lactation, 240 days, respect- ively). Three different datasets were prepared, where rumin- ation, milk, and activity were averaged as means of 1, 3, and 5 days before the response variable record. On each dataset, two models were fitted: logistic regression and random forest. All the analyses were performed in R using the caret package. Data were divided into a training and a testing dataset (pro- portion 80/20). Training data was used to train two different algorithms which were used to predict the class variable. The two selected algorithms were: 1. a logistic regression 2. a ran- dom forest. For all the datasets, logistic regression was not able to predict diseased individuals, assigning all to the 'healthy' class. Random forest performed better, although with a high-class error. The 5-day window had the lowest OOB error rate (0.24%) and the lowest class error (0.71). Further tuning of the selected models will be necessary to build a valuable tool to predict locomotion system issues.

This work describes the activities planned by CINI and CNR-ISTI for the urban innovation project of the neighborhood "I passi" in Pisa. Specifically, these activities fit three lines of intervention concerning the energy monitoring of a new co-housing building, measures for the assistance of elderly living in this co-housing structure, and a participatory sensing platform for the residents of the neighborhood.

The Wireless Network Laboratory at ISTI-CNR studies how to exploit mobility and sociality of people in a mobile social network in order to advertise and discover the services provided by devices. This problem is commonly referred to as service discovery. An efficient strategy for advertising to other devices about the existence of new or existing services is proposed alongside a strategy for discovering a specific service.

The Italian Goat Consortium (IGC), joined the effort of many Universities and Research Institutes, in a comprehensive study of the Italian goat population genetic makeup using a medium density (54K) SNPs chip. Currently IGC has geno- typed more than 1,000 animals from more than 30 goat breeds and populations from all Italian geographical and agro- ecological areas of goat rearing. The aim of this work is to obtain a clear picture of the Italian caprine biodiversity, to reconstruct the ancestry, to disentangle the genetic background and to assess the relationships among and within the investigated breeds. To date, the IGC dataset includes about 50 million genotypes. The data were quality checked by excluding markers and individuals on the basis of missing genotypes, minor allele frequency and close individual relatedness. Genetic relationships among and within breeds was investigated by Multi-Dimensional Scaling and Principal Component Analysis. Population structure, ancestry models and admixture were estimated by ADMIXTURE and fastSTRUCTURE software. Finally, phylogenic trees were recon- structed with PHYLIP software suite starting from shared-allele identity by state, and Reynolds distance matrices, while past migration events were modeled with TreeMix software. The results confirmed high levels of genetic polymorphism and confirmed the North-South geographical pattern of diver- sity, previously reported on a smaller sample of Italian goat breeds. The analysis also revealed a pivotal role of Central Italy in connecting the genetic resources of the northern and southern areas of the country, and confirms the genetic isola- tion of insular breeds. Moreover, some breeds show clearly distinctive and homogeneous gene pools, whereas other breeds present complex and, in some cases, dishomogeneous genetic background. Even if "A breed is a group of domestic animals, termed such by common consent of the breeders" (Lush J.L., 1994), genomic tools are useful in understanding the genetic back- ground of populations and in defining their relationships or uniqueness. These tools can complement the traditional ones in providing farmers and their associations a powerful aid for a more conscious management of goat populations and their biodiversity.

Negli ultimi anni l’attenzione del consumatore si è sempre più orientata verso le caratteristiche nutrizionali degli alimenti. Queste proprietà sono di grande importanza anche per quanto riguarda le produzioni dei piccoli ruminanti. Il presente lavoro ha lo scopo di riassumere i principali risultati emersi dal progetto di ricerca “Aspetti della produzione dei piccoli ruminanti con particolare impatto sulla salute umana”. Sono stati analizzati mediante i metodi descritti in letteratura: 1) i polimorfismi genetici dei biopeptidi del latte dei piccoli ruminanti; 2) le attività di alcuni enzimi della membrana del globulo di grasso e la frazione lipidica del latte ovino; 3) la qualità nutrizionale del latte e del formaggio ovino in relazione all’intensità di pascolamento; 4) le componenti bioattive di siero e scotta residui alla produzione dei formaggi ovi-caprini; 5) la resistenza genetica alle encefalopatie spongiformi trasmissibili e l’efficienza economica e biologica in razze ovine. I risultati ottenuti evidenziano, da svariati punti di vista, numerose potenzialità legate alle produzioni dei piccoli ruminanti e alle loro ricadute sulla salute umana.

Deliverable D4.3 describes the process of training, model selection and validation adopted to identify the most adequate setting (and fitting of the parameters) for the activity recognition components developed in WP4 and deployed in the DOREMI pilot phase. A detailed account of the experimental assessment of the expected performances on DOREMI laboratory data collected by task 6.1 (as per DoW) is provided. The deliverable also reports basic aggregated statistics on the data generated by the activity recognition and reasoning components during the DOREMI pilot experimental phases. Finally, it is provided a post-hoc assessment of the unsupervised exploratory data analysis and reasoning components following the UK pilots.

L'obiettivo di questo tirocinio è l'implementazione di un modulo di comunicazione multipath per il caso d'uso di trasmissione di flussi multimediali in real-time. Si definisce multipath una connessione in grado di sfruttare contemporaneamente piu` canali fisici di comunicazione fra sorgente e destinazione. Il modulo ha l'obiettivo principale di gestire la distribuzione del traffico sui diversi canali tramite un opportuno algoritmo di scheduling. Nella fase iniziale, il lavoro si è concentrato sulla ricerca dei protocolli multipath esistenti, nonchè delle piattaforme software in grado di soddisfare tale requisito. Sono stati dapprima individuati tre protocolli in grado di gestire comunicazioni multipath: . SCTP (Stream Control Transmission Protocol), descritto nel capitolo 1; . MP-TCP (Multipath-TCP), descritto nel capitolo 2; . MP-RTP (Multipath-RTP), descritto nel capitolo 3. Tale studio ha portato, infine, alla scelta del protocollo piu` adatto al caso d'uso proposto, ovvero MPRTP. Nella seconda fase, è stata selezionata la piattaforma software di riferimento, ovvero Gstreamer1, descritta nel capitolo 4.1, per la quale è presente un plugin che implementa il protocollo MPRTP. Su tale plugin - ancora in via di sviluppo - è stato implementato un algoritmo di scheduling che gestisca la distribuzione del traffico multimediale sui flussi multipath, in accordo con le statistiche di ciascun canale.

Introduction: European population aging requires the design of innovative solutions able to support, in term of quality and time, its health status. DOREMI project has developed an innovative platform able to stimulate and monitor elder people as also to be customized on user requirements. Methods: 34 older people (age 65-80 years) were involved in UK and Italy DOREMI trials (3 months). Subjects were characterized at baseline in terms of physical activity (SPPB, PASE test, daily steps/ meters, 6MWT), hemodynamic and biochemical parameters (blood pressure, HR, lipid profile, glycaemia, etc.), dietary habits (caloric intake) and balance assessment (BERG scale). Through the DOREMI technological platform, users were: stimulated to perform indoor physical activity protocol (exergame on tablet); monitored by DOREMI bracelet (heart rate monitoring); invited to fill diet e-diary, receiving nutritional advice provided by the expert through the same application; tested for balance by item-10 BERG test (DOREMI balance board). At the end of trial, users underwent the same test battery of baseline. Results: In DOREMI population an overall increase in physical activity was observed, with a significant improvement in hemodynamic (decrease in blood pressure and HR at 6MWT) and biochemical parameters (decrease in LDL, triglycerides, total cholesterol). The overall effect of dietary advice and physical activity protocol on subject's balance is under investigation. Key conclusions: An integrated approach of physical activity and diet, supported by DOREMI technological platform, represents an innovative approach to stimulate healthy and active ageing of population, with a potential cost-reduction for European health care systems in middle-long period.

Paired pulse facilitation (PPF) is a form of short-term synaptic plasticity that results from an interaction of residual presynaptic Ca(2+) ([Ca(2+)](res)), number of release-competent vesicles, and the sensitivity of the vesicle release mechanisms to Ca(2+). While PPF is predominant at hippocampal Schaffer collateral-CA1 (SC-CA1) synapses, facilitation is greater in adult mice (designated Tkneo) that over express an isoform of the plasma membrane-targeted SNARE protein, SNAP-25a, which is normally predominantly expressed in juvenile animals. SNAP-25 is essential for action potential-dependent neuroexocytosis, yet the significance of the shift between the alternatively spliced variants SNAP-25a and SNAP-25b is not fully understood. This alteration of a key component of the protein machinery required for neurotransmitter release in Tkneo mice, therefore, provides a useful tool to further investigate presynaptic mechanisms that influence short-term plasticity. To explore this link between SNAP-25 and PPF, we simultaneously measured postsynaptic potentials and presynaptic [Ca(2+)](res) during paired-pulses in adult Tkneo, heterozygote null (HET), and wild type (WT) mice. We demonstrate that enhanced PPF is maintained at mature hippocampal synapses of Tkneo mice that predominantly express SNAP-25a, and that [Ca(2+)](res) kinetics are altered at synapses of Tkneo and HET mice, both of which exhibit reduced levels of total SNAP-25 expression. To evaluate the role of SNAP-25 in short-term plasticity and [Ca(2+)](res) regulation, we applied a vesicular release probability model for neurotransmission. Our results suggest that the isoform expression and total level of SNAP-25 affect both [Ca(2+)](res) dynamics and the ability of releasable vesicles to enter into a facilitated state.

Presynaptic Ca(2+) influx pathways, cytoplasmic Ca(2+) buffering proteins and Ca(2+) extrusion processes undergo considerable change during the first postnatal month in rodent neurons. These changes may be critical in establishing short-term plasticity at maturing presynaptic terminals where neurotransmitter release is directly dependent on the dynamics of cytoplasmic residual Ca(2+) ([Ca(2+)](res)). In particular, the robust paired-pulse facilitation characteristic of adult neurons is almost entirely lacking in newborns. To examine developmental changes in processes controlling [Ca(2+)](res), we measured the timecourse of [Ca(2+)](res) decay in presynaptic terminals of Schaffer collateral to CA1 synapses in acute hippocampal slices following single and paired orthodromic stimuli in the stratum radiatum. Developmental changes were observed in both the rise time and slow exponential decay components of the response to single stimuli such that this decay was larger and faster in the adult. Furthermore, we observed a greater caffeine-sensitive basal Ca(2+) store, which was differentially affected when active uptake into the endoplasmic reticulum was blocked, in the presynaptic fields of the Schaffer collateral to CA1 terminals of P6 and younger mice when compared to adults. These transitions in [Ca(2+)](res) dynamics occurred gradually over the first weeks of postnatal life and correlated with changes in short-term plasticity.

All countries in Europe are experiencing an ageing of their populations, with a decrease in the number of people of working age per retiree. Health trends among the elderly are mixed: severe disability is declining in some countries but increasing in others, while mild disability and chronic disease are generally increasing. As a consequence, long-term care costs are certain to increase with the ageing of the population, unless appropriate measures are implemented and elderly people empowered to follow them. According to the University College Dublin Institute of Food and Health, three are the most notable health promotion and disease prevention programs that target the main causes of morbidity and premature mortality: malnutrition, sedentariness, and cognitive decline, conditions that affect the quality of life of elderly people and drive to disease progression. These three features represent the target areas in the DOREMI (Decrease of cOgnitive decline, malnutRition and sedEntariness by elderly empowerment in lifestyle Management and social Inclusion) Project. The project aims at developing a systemic solution for elderly, able to prolong the functional and cognitive capacity by empowering, stimulating and unobtrusively monitoring the daily activities according to well-defined "Active Ageing" lifestyle protocols. The project joins the concept of prevention centered on the elderly, characterized by an unified vision of being elderly today, namely, a promotion of the health by a constructive interaction among mind, body and social engagement.