Search

Uncertainty in dense heterogeneous IoT sensor networks can be decreased by applying reputation-inspired algorithms, such as the EWMA (Exponentially Weighted Moving Average) algorithm, which is widely used in social networks. Despite its popularity, the eventual convergence of this algorithm for the purpose of IoT networks has not been widely studied, and results of simulations are often taken in lieu of the more rigorous proof. Therefore the question remains, whether under stable conditions, in realistic situations found in IoT networks, this algorithm indeed converges. This paper demonstrates proof of the eventual convergence of the EWMA algorithm. The proof consists of two steps: it models the sensor network as the UOG (Uniform Opinion Graph) that enables the analytical approach to the problem, and then offers the mathematical proof of eventual convergence, using formalizations identified in the previous step. The paper demonstrates that the EWMA algorithm converges under all realistic conditions.

The growing popularity of inexpensive IoT (Internet of Things) sensor networks makes their uncertainty an important aspect of their adoption. The uncertainty determines their fitness for purpose, their perceived quality and the usefulness of information they provide. Nevertheless, neither the theory nor the industrial practice of uncertainty offer a coherent answer on how to address uncertainty of networks of this type and their components. The primary objective of this paper is to facilitate the discussion of what progress should be made regarding the theory and the practice of uncertainty of IoT sensor networks to satisfy current needs. This paper provides a structured overview of uncertainty, specifically focusing on IoT sensor networks. It positions IoT sensor networks as contrasted with professional measurement and control networks and presents their conceptual sociotechnical reference model. The reference model advises on the taxonomy of uncertainty proposed in this paper that demonstrates semantic differences between various views on uncertainty. This model also allows for identifying key challenges that should be addressed to improve the theory and practice of uncertainty in IoT sensor networks.

The aim of this paper is to introduce a NUT model (NUT: network-uncertainty-trust) that aids the decrease of the uncertainty of measurements in autonomous hybrid Internet of Things sensor networks. The problem of uncertainty in such networks is a consequence of various operating conditions and varied quality of measurement nodes, making statistical approach less successful. This paper presents a model for decreasing the uncertainty through the use of socially inspired metaphors of reputation, trust, and confidence that are the untapped latent information. The model described in the paper shows how the individual reputation of each node can be assessed on the basis of opinions provided by other nodes of the hybrid measurement network, and that this method allows to assess the extent of uncertainty the node introduces to the network. This, in turn, allows nodes of low uncertainty to have a greater impact on the reconstruction of values. The verification of the model, as well as examples of its applicability to air quality measurements are presented as well. Simulations demonstrate that the use of the model can decrease the uncertainty by up to 55% while using the EWMA (exponentially weighted moving average) algorithm, as compared to the reference one.

The objective of this study was to estimate the relationships between cadmium (Cd) levels in the small intestine and other organs (kidney, liver, lungs) and factors influencing the intestinal Cd levels in humans, as based on autopsy analysis of subjects not exposed to Cd occupationally. The study also involved estimating the levels of zinc (Zn) in these organs, as it is known that this element exerts interactions with Cd at the level of absorption and tissue binding. The levels of Cd and Zn were determined in the renal cortex, liver, lungs and three fragments of the small intestine (duodenum, jejunum, ileum) of 29 subjects deceased at the age 42 +/- 13 years. Flame atomic absorption spectrometry (AAS; kidneys, liver) and flameless AAS (lungs, intestine) were used. The level of Cd in the lungs was used as a marker of smoking habit. The determined levels (mean +/- SD) were: 0.28 +/- 0.16 microg Cd/g and 15.2 +/- 3.4 microg Zn/g in the duodenum; 0.26 +/- 0.15 microg Cd/g and 16.9 +/- 3.7 microg Zn/g in the jejunum; 0.13 +/- 0.07 microg Cd/g and 14.6 +/- 5.4 microg Zn/g in the ileum. Intestinal Cd levels are correlated with organ and total body Cd, and this was best expressed for Cd in ileum (r=0.67 with renal, r=0.71 with hepatic and r=0.68 with total Cd). In conclusions, the levels of Cd in the small intestine of humans are relatively low and reflect predominantly the whole body retention of this element. Somewhat higher levels of Cd are contained in the initial parts of the small intestines. In all fragments of small intestines the levels of Cd are higher in smokers. Also, the levels of Zn were relatively low and did not correlate with the levels of Cd.

This article is based on data on the levels of metals (Cd, Zn, Cu) and metallothionein (MT) determined radiochemically with 203Hg in renal cortex and liver of 137 autopsy cases. From this number, for 23 cases, the gel filtration of the cytoplasmic fraction of the organs was performed. The molar content of metals in the MT fraction (Sephadex G-50) amounted to 46.9, 50.2, and 2.0% for Cd, Zn, and Cu in renal cortex, respectively, and to 8.3, 83.6, and 9.1% for Cd, Zn, and Cu in the liver, respectively. In parallel with the increase of Cd and MT in renal cortex, increasing saturation was found of the MT fraction by Cd, occurring at the expense of Zn and Cu. Equimolar amounts of Cd and Zn in the MT fraction are found at Cd level of 0.5 micromol Cd/g wet wt of renal cortex. In the liver, analogous dependency (elevation of %Zn, depression of %Cd and %Cu) were observed in relation to Zn and MT levels in this organ. The basic level of Zn (not bound with MT) was estimated at 0.5 micromol/g for both renal cortex and liver. A deficit of non-MT Zn in kidneys is proposed as an alternative mechanism of toxic Cd action.

Tissues were analysed from 60 subjects deceased at the age of 18-82 (mean 49 years), inhabitants of the southwestern region of Poland (Legnica Copper Basin). Mean cadmium, zinc and copper concentrations determined by flame AAS were for renal cortex 41.8; 47.8 and 2.3 micrograms Me/g tissue, respectively. The respective levels in the liver were 2.3; 51.6 and 4.6 micrograms Me/g tissue (w.w.). A broad range of cadmium levels was found in renal cortex (4.2-129.3.micrograms/g), with the highest values in the 40-60 age group. The effect of tobacco smoking is more evident than in other countries: 26.3 micrograms/g in non-smokers vs. 54.6 micrograms/g in smokers. Similar proportions were found by computing the whole body burden of cadmium. Compared with other regions of Poland the environmental exposure of humans to cadmium is moderate, however, it is higher than in other European countries.

The levels of Cd, Zn, and Cu in the kidneys and liver of 23 pigeons from Lodz and Cracow were determined by flame AAS, after wet digestion. Metallothionein (MT) was determined radiochemically using 203Hg, in the liver of 5 pigeons. The mean concentrations were: 7.2 micrograms Cd/g; 43.2 micrograms Zn/g and 5: 3 micrograms Cu/g w.w in the kidneys; and 1, 4 micrograms Cd/g; 51.5 micrograms Zn/g and 5.0 micrograms Cu/g w.w in the liver. The correlations between Cd in kidneys and liver (r = 0.88), and MT (mean: 0.37 mumol Hg/g w.w) vs Zn (r = 0.99) in the liver were found. The results indicate higher environmental pollution by cadmium in Cracow (kidneys: 11.7 micrograms Cd/g (p < 0.06); liver: 2.1 micrograms Cd/g (p < 0.01) than in Lodz (4.4 and 1.0 micrograms Cd/g respectively), and are in accordance with preliminary human data in the inhabitants of these two regions. It can be assumed that pigeons are a sensitive biomonitor of environmental pollution by cadmium.

Cd, Zn, Cu, and metallothionein (MT) levels have been determined in the renal cortex and liver of 70 persons who died in Lodz and its surroundings in the years 1985-1989. The mean concentrations were: 44.9±28.6 µg Cd/g, 52.0±16.7 µg Zn/g, 2.4±1.0 µg Cu/g, 0.79±0.40 µmol Hg/g, and 3.5±1.8 µg Cd/g, 66.7±30.5 µg Zn/g, 4.9±2.1 µg Cu/g, 0.50±0.38 µmol Hg/g wet tissue in renal cortex and liver, respectively, with mean age 54.0±13.8. Smokers showed 2.4 times higher levels of Cd in the renal cortex than non-smokers. The mean body burden of Cd was 33.4±17.3 mg. Smoking increases it twofold from 22.0 mg in non-smokers to 41.8 mg in smokers.