Automatic Cleaning of Time Series Data in Rural Internet of Things Ecosystems That Use Nomadic Gateways.

Sensors (Basel)

Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Narutowicza 11/12, 80-233 Gdańsk, Poland.

Published: January 2025

A serious limitation to the deployment of IoT solutions in rural areas may be the lack of available telecommunications infrastructure enabling the continuous collection of measurement data. A nomadic computing system, using a UAV carrying an on-board gateway, can handle this; it leads, however, to a number of technical challenges. One is the intermittent collection of data from ground sensors governed by weather conditions for the UAV measurement missions. Therefore, each sensor should be equipped with software that allows for the cleaning of collected data before transmission to the fly-over nomadic gateway from erroneous, misleading, or otherwise redundant data-to minimize their volume and fit them in the limited transmission window. This task, however, may be a barrier for end devices constrained in several ways, such as limited energy reserve, insufficient computational capability of their MCUs, and short transmission range of their RAT modules. In this paper, a comprehensive approach to these problems is proposed, which enables the implementation of an anomaly detector in time series data with low computational demand. The proposed solution uses the analysis of the physics of the measured signals and is based on a simple anomaly model whose parameters can be optimized using popular AI techniques. It was validated during a full 10-month vegetation period in a real Rural IoT system deployed by Gdańsk Tech.

Download full-text PDF

Source
http://dx.doi.org/10.3390/s25010189DOI Listing

Publication Analysis

Top Keywords

time series
8
series data
8
data
5
automatic cleaning
4
cleaning time
4
data rural
4
rural internet
4
internet things
4
things ecosystems
4
ecosystems nomadic
4

Similar Publications

Deciphering the colostral-immunity transfer: from mammary gland to neonates small intestine.

Vet Res Commun

January 2025

Laboratory of Agricultural Animal Genetics, Breeding and Reproduction of Ministry of Education, Huazhong Agricultural University, Wuhan, China.

Colostrum, the initial mammary secretion produced by various mammals following birth, is a conduit for maternal immunity transfer in diverse mammalian species. Concurrently, many cellular processes are occurring in the neonatal small intestine to prepare it to receive molecular signals from a superfood essential for the neonate's health and development. During the prepartum colostrum secretion, the newborn intestine undergoes transient alterations in the intestinal barrier, primarily regulating immunoglobulin absorption.

View Article and Find Full Text PDF

Introduction: The objective of this study is to compare the 5 year overall survival of patients with stage I-III colon cancer treated by laparoscopic colectomy versus open colectomy.

Methods: Using Mecklenburg-Western Pomerania Cancer Registry data from 2008 to 2018, we will emulate a phase III, multicenter, open-label, two-parallel-arm hypothetical target trial in adult patients with stage I-III colon cancer who received laparoscopic or open colectomy as an elective treatment. An inverse-probability weighted Royston‒Parmar parametric survival model (RPpsm) will be used to estimate the hazard ratio of laparoscopic versus open surgery after confounding factors are balanced between the two treatment arms.

View Article and Find Full Text PDF

Managerial factors affecting milking-abilities of Holstein cattle under intensive production system in Egypt.

Trop Anim Health Prod

January 2025

Department of Animal Production, Faculty of Agriculture, Menoufia University, Shibin Al Kawm, Egypt.

This article aims to explore milking-ability criteria of Holstein dairy cattle under intensive production system in Egypt and investigate some managerial factors that influence them in dairy farms. The data obtained from five herds belong to a commercial intensive production system farm, Egypt. Data included 3509 records.

View Article and Find Full Text PDF

Introduction: We sought to explore the variability of antibody responses to multiple vaccines during early life in individual children, assess the trajectory of each child longitudinally, determine the associations of demographic variables and antibiotic exposures with vaccine-induced immunity, and link vaccine responsiveness to infection proneness.

Methods: In 357 prospectively-recruited children, age six through 36 months, antibody levels to 13 routine vaccine antigens were measured in sera at multiple time points and normalized to their respective protective thresholds to categorize children into four groups: very low, low, normal, and high vaccine responder. Demographic variables and frequency of antibiotic exposure data were collected.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!