A data estimation for failing nodes using fuzzy logic with integrated microcontroller in wireless sensor networks

Continuous data transmission in wireless sensor networks (WSNs) is one of the most important characteristics which makes sensors prone to failure. a backup strategy needs to co-exist with the infrastructure of the network to assure that no data is missing. The proposed system relies on a backup strategy of building a history file that stores all collected data from these nodes. This file is used later on by fuzzy logic to estimate missing data in case of failure. An easily programmable microcontroller unit is equipped with a data storage mechanism used as cost worthy storage media for these data. An error in estimation is calculated constantly and used for updating a reference “optimal table” that is used in the estimation of missing data. The error values also assure that the system doesn’t go into an incremental error state. This paper presents a system integrated of optimal data table, microcontroller, and fuzzy logic to estimate missing data of failing sensors. The adapted approach is guided by the minimum error calculated from previously collected data. Experimental findings show that the system has great potentials of continuing to function with a failing node, with very low processing capabilities and storage requirements.

Research Note on Captive Atlantic Flows: Estimating Missing Data by Slave-Voyage Routes

This essay provides new estimates of the number of captives carried in the Atlantic slave trade during each decade from the 1650s to the 1860s. It relies on two categories of known data—on the routes of voyages and the numbers of captives recorded on those voyages—as a basis for estimation of missing data and totals of captive flows. It uses techniques of Bayesian statistics to estimate missing data on routes and flows of captives. As a framework for the Bayesian estimates, it focuses on analysis of 40 distinct routes linking the African coast to the Americas and traces the captive flows—that is, the number of captives embarked on or disembarked from voyages along those routes. The dataset that provides the basis for this research note is available at: https://doi.org/10.7910/DVN/6HLXO3.

Synthesizing test Reference Year files from known climate patterns of nearby cities

The importance of an architecture adapted to its climatic context is often debated. In order to avoid future unexpected environmental behavior or failure of a building during its use, building simulation tools are used in the design and require complete and consistent weather data. However, such data are not always available for the locations where buildings are simulated, and the use of data from neighboring cities becomes usual. There are, though, several uncertainties involved in the behavior of environmental variables when the climate of large urban centers is attributed to nearby localities and areas with more significant vegetation cover, water bodies, different topography, among others. The present paper aims to present the process of preparing a weather file for the Pecém Industrial and Port Complex, located at 40 km from the capital Fortaleza, Brazil, in order to be used in simulations during the design process of buildings. The synthesis of the file was achieved through the collection and treatment of information measured in loco, the application of recommended models for the estimation of missing data, and the development of an alternative method for the estimation of a Test Reference Year of localities without weather data of several years.

Data-driven missing data imputation in cluster monitoring system based on deep neural network

Due to cluster instability, not in the cluster monitoring system. This paper focuses on the missing data imputation processing for the cluster monitoring application and proposes a new hybrid multiple imputation framework. This new imputation approach is different from the conventional multiple imputation technologies in the fact that it attempts to impute the missing data for an arbitrary missing pattern with a model-based and data-driven combination architecture. Essentially, the deep neural network, as the data model, extracts deep features from the data and deep features are further calculated then by a regression or data-driven strategies and used to create the estimation of missing data with the arbitrary missing pattern. This paper gives evidence that if we can train a deep neural network to construct the deep features of the data, imputation based on deep features is better than that directly on the original data. In the experiments, we compare the proposed method with other conventional multiple imputation approaches for varying missing data patterns, missing ratios, and different datasets including real cluster data. The result illustrates that when data encounters larger missing ratio and various missing patterns, the proposed algorithm has the ability to achieve more accurate and stable imputation performance.