TASK Quarterly

Analysis of cores affinity within the containerized environment based on selected IOT middleware - observations and recommendations

2024-04-04T08:09:17+02:00

The Internet of Things gets bigger and bigger audiences. This topic is really popular in science and also in industry. There are many fields for research. One of them is efficient deployment against resource utilization. Another one is containerization within IoT platforms. One of the commonalities of these two topics is different CPU affinity against containerized platforms to get the best performance. There were plenty of papers dedicated to containerization even in IoT but none of these focused on core affinity. As this survey analyzes the scalability and stability of the platform in different core-container configurations based on the IoT platform - DeviceHive, it brings a novelty to this area. Most interesting observations were made in the field of the same configurations in terms of the number of nodes but varying with core affinity. Analyzed observations may be useful during the architecture planning phase for containerized IoT platforms.

Application of artificial neural networks in the development of the PM10 air pollution prediction system

2024-04-04T08:19:27+02:00

This article presents research on the model of forecasting the average daily air pollution levels focused mainly on two solutions, artificial neural networks: the NARX model and the LSTM model. The research used an air quality monitoring system. This system includes individually designed and implemented sensors to measure the concentration of pollutants such as PM10, PM2.5, SO₂, NO₂ and to record weather conditions such as temperature, humidity, pressure, wind strength and speed. Data is sent to a central database server based on the MQTT protocol. Additional weather information in the area covered by pollution monitoring is collected from the weather services of the IMGW and openwethermap.org. The artificial neural network models were built in the MATLAB environment, the process of learning neural networks was performed and the results of pollution prediction for the level of PM10 dust were tested. The models showed good and acceptable results when forecasting the state of PM10 dust concentration in the next 24 hours. The LSTM prediction model were more accurate than the NARX model. The future work will be related to the use of artificial intelligence algorithms to predict the concentration of other harmful substances, e.g. PM2.5, NO₂, SO₂ etc. A very important task in the future will be to frame the entire system of monitoring and predicting smog in a given area.

Three dimensional visualization of histopathological data

2024-04-05T07:36:24+02:00

The histopathological examination provides information about the spatial assessment of pathological changes in the tissue. The authors present a method of extending this histopathological spatial assessment with a 3D view consisting of images of microscopic layers. The proposed solution creates 3D models based on images obtained from the database of the Medical University of Gdańsk (Digital Pathology). First, a series of medical images related to the study of a specific pathological tissue undergoes a process of background detection and removal through an algorithm. Next, images aligned with each other. Then, two types of 3D models are created: 1) classical model with Marching Cubes algorithm and 2) the use Cloud of Points.