Wyniki wyszukiwania - BazTech

1

The impact of macroeconomic factors on the profitability of companies. Analysis using neural networks

Szutowski Dawid

Zeszyty Naukowe. Organizacja i Zarządzanie / Politechnika Śląska

|

2024

|

z. 193

667--683

EN

Purpose: Despite the central role of profitability in economic analysis, previous research has yielded diverse and often unstructured conclusions regarding its determinants. To address this gap, this empirical investigation aimed to explore the major determinants of company profitability. Design/methodology/approach: It conducted a comprehensive analysis of factors, encompassing: changes in the gross domestic product, Consumer Price Index, Producer Price Index, NBP’s Reference rate, investment outlays, intramural expenditures on research and development, expenditures on innovation activities in enterprises, and patents granted, alongside company-level profitability indicators. The study's sample consisted of companies representing 19 sectors of the economy, spanning from 2004 to 2021. For data analysis, a neural network was employed, specifically a multi-layer perceptron (MLP) utilizing the sigmoid activation function. Findings: The findings suggest that alterations in macroeconomic variables can significantly impact the profitability of companies. The analysis carried out revealed that consumer price index, reference rate, gross domestic product and producer price index were the most important exogeneous factors. Originality/value: This study introduces several novelties, including the application of neural networks, which are infrequently utilized in this field, and the simultaneous analysis of a comprehensive set of independent variables.

2

Analysis of the possibility of increasing the level of reliability of Fire Alarms Systems through integration with Video-Based Flame Detection System

Mazur Michał, Wiśnios Michał, Paś Jacek

Journal of KONBiN

|

2024

|

Vol. 54, iss. 2

71--91

EN

Classic Fire Alarm Systems (FAS) are characterized by a high level of false alarms in relation to the number of confirmed reports. To increase the level of reliability of operation of this type of system, it was proposed to integrate it with a Video-Based Flame Detection System (VBFDS). For this purpose, a video-based fire detection algorithm was designed. In addition, methods popular in the literature for increasing the reliability of such systems, such as color filtering, the Tracking Growth Object (TGO) factor, and the use of the Naive-Bayes (NB) classifier, were tested. The purpose of the article is to analyze these methods in comparison with the basic version of the algorithm, as well as the possibility of integrating Video-Based Flame Detection System (VBFDS) with classic FAS.

PL

Klasyczne Systemy Sygnalizacji Pożaru (SSP) charakteryzują się wysokim poziomem fałszywych alarmów w stosunku do liczby potwierdzonych zgłoszeń. Aby zwiększyć poziom niezawodności działania tego typu systemów, zaproponowano jego integrację z Systemem Wizyjnej Detekcji Płomienia (SWDP). W tym celu zaprojektowany został algorytm wizyjnej detekcji pożaru. Dodatkowo przetestowane zostały popularne w literaturze metody zwiększenia niezawodności tego typu systemów, takie jak filtracja kolorów, współczynnik TGO (ang. Tracking Growth Object) oraz zastosowanie klasyfikatora NB (ang. Naive-Bayes). Celem artykułu jest analiza tych metod w porównaniu z podstawową wersją algorytmu oraz możliwość integracji SWDP z klasycznymi SSP.

3

Smart non-destractive test of a concrete wall using a hammer

Ito Atsushi, Koike Masafumi, Hibino Katsuhiko

International Journal of Electronics and Telecommunications

|

2024

|

Vol. 70, No. 1

45--50

EN

Large concrete structures such as buildings, bridges, and tunnels are aging. In Japan and many other countries, those built during economic reconstruction after World War II are about 60 to 70 years old, and flacking and other problems are becoming more noticeable. Periodic inspections were made mandatory by government and ministerial ordinance during the 2013-2014 fiscal year, and inspections based on the new standards have just begun. There are various methods to check the soundness of concrete, but the hammering test is widely used because it does not require special equipment. However, long experience is required to master the hammering test. Therefore, mechanization is desired. Although the difference between the sound of a defective part and a normal part is very small, we have shown that neural network is useful in our research. To use this technology in the actual field, it is necessary to meet the forms of concrete structures in various conditions. For example, flacking in concrete exists at various depths, and it is impossible to learn about flacking in all cases. This paper presents the results of a study of the possibility of finding flacking at different depths with a single inspection learning model and an idea to increase the accuracy of a learning model when we use a rolling hammer.

4

Application of Resnet-152 neural networks to analyze images from UAV for fire detection

Stelmakh Nataliia, Mandrovska Svitlana, Galagan Roman

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2024

|

T. 14, nr 2

77--82

EN

Timely detection of fires in the natural environment (including fires on agricultural land) is an urgent task, as their uncontrolled development can cause significant damage. Today, the main approaches to fire detection are human visual analysis of real-time video stream from unmanned aerial vehicles or satellite image analysis. The first approach does not allow automating the fire detection process and contains a human factor, and the second approach does not allow detect the fire in real time. The article is devoted to the issue of the relevance of using neural networks to recognize and detect seat of the fire based on the analysis of images obtained in real time from the cameras of small unmanned aerial vehicles. This ensures the automation of fire detection, increases the efficiency of this process, and provides a rapid response to fires occurrence, which reduces their destructive consequences. In this paper, we propose to use the convolutional neural network ResNet-152. In order to test the performance of the trained neural network model, we specifically used a limited test dataset with characteristics that differ significantly from the training and validation dataset. Thus, the trained neural network was placed in deliberately difficult working conditions. At the same time, we achieved a Precision of 84.6%, Accuracy of 91% and Recall of 97.8%.

PL

Wczesne wykrycie pożarów w środowisku naturalnym (w tym pożarów na gruntach rolnych) jest zadaniem pilnym, gdyż ich niekontrolowany rozwój może spowodować znaczne szkody. Obecnie głównymi podejściami do wykrywania pożarów jest wizualna analiza przez człowieka strumienia wideo w czasie rzeczywistym z bezzałogowych statków powietrznych lub analiza obrazu satelitarnego. Pierwsze podejście nie pozwala na automatyzację procesu wykrywania pożaru i uwzględnia czynnik ludzki, natomiast drugie podejście nie pozwala na wykrycie pożaru w czasie rzeczywistym. Artykuł poświęcony jest zagadnieniu przydatności wykorzystania sieci neuronowych do rozpoznawania i wykrywania źródła pożaru na podstawie analizy obrazów uzyskiwanych w czasie rzeczywistym z kamer małych bezzałogowych statków powietrznych. Zapewnia to automatyzację wykrywania pożaru, zwiększa efektywność tego procesu oraz zapewnia szybką reakcję na wystąpienie pożarów, co ogranicza ich niszczycielskie skutki. W artykule proponujemy wykorzystanie splotowej sieci neuronowej ResNet-152. Aby przetestować wydajność wyszkolonego modelu sieci neuronowej wykorzystaliśmy ograniczony testowy zbiór danych, którego charakterystyka znacznie różni się od zbiorów danych treningowych i walidacyjnych. Tym samym wytrenowana sieć neuronowa została poddana celowo trudnym warunkom operacyjnym. Jednocześnie uzyskano parametry "Precision" – 84.6%, "Accuracy" – 91% i "Recall" – 97.8%.

5

Building Semantic Segmentation Using UNet Convolutional Network on SpaceNet Public Data Sets for Monitoring Surrounding Area of Chan Chan (Peru)

Chicchon Miguel, Malinverni Eva Savina, Sanità Marsia, Pierdicca Roberto, Colosi Francesca, Trujillo Francisco James León

Geomatics and Environmental Engineering

|

2024

|

Vol. 18, no. 3

25--43

EN

The amount of damage to cultural heritage sites is increasing rapidly every year. This is due to inadequate heritage management and uncontrolled urban growth as well as unpredictable seismic and atmospheric events that manifest themselves in a continuously deteriorating ecosystem. Thus, applications of artificial intelligence (AI) in remote-sensing (RS) techniques (machine-learning and deep-learning algorithms) for monitoring archaeological sites have increased in recent years. This research involves the surrounding area of the archaeological site of Chan Chan in Peru in particular. An approach that is based on the use of AI algorithms for building footprint segmentation and changedetection analysis by means of RS images is proposed. It involves a UNet convolutional network based on an EfficientNet B0 to B7 encoder. The network was trained on two public data sets from SpaceNet that were based on WV2 and WV3 satellite images: SpaceNet V1 (Rio), and SpaceNet V2 (Shanghai). In the pre-processing phase, the images from the two data sets have been equalized in order to improve their quality and avoid overfitting. The building segmentation has been performed on HRV images of the study area that were downloaded from Google Earth Pro. The value that was achieved in the IoU metric was around 70% in both experiments. The purpose of this proposed methodology is to assist scientists in drafting monitoring and conservation protocols based on already-recorded data in order to prevent future disasters and hazards.

6

Total phosphorus and total Kjeldahl nitrogen removal using an aerobic granular sludge process. Case studies ANN and RSM modeling

Sanchez-Sanchez Celina, Flores-Gomez Jean, Moeller-Chávez Gabriela, Morales-Rivera Juan

Environment Protection Engineering

|

2024

|

Vol. 50, nr 1

123--137

EN

Removing nutrients from wastewater is essential because high concentrations in aquatic systems lead to severe eutrophication problems, the most common impairment of surface waters such as lakes and oceans. Total phosphorus (TP) and total Kjeldahl nitrogen (TKN) were removed from mixed wastewater using an aerobic granular sludge process in a sequencing batch reactor (AGS-SBR). An artificial neural network (ANN) and response surface methodology (RSM) were applied to evaluate the main parameters of the process. For TKN removal, only cycle time (CT) (0.0475) was a significant variable, achieving removal efficiencies of up to 81%. In TP case removal, two parameters, VER and AR, were substantial for this process, completing elimination efficiencies of around 40%. On comparing the models with statistical indices, ANN coupled with the moth-flame optimization algorithm (ANN-MFO) demonstrated higher performance with an adjusted R2 (0.9866) for the case of TP removal and (0.9519) for TKN removal.

7

Prediction of groundwater contamination in an open landfill area using a novel hybrid clustering-based AI model

Singh Prachi, Mahor Vikas, Lakshmaiya Natrayan, Shanker Kandukuri, Kaliappan Seeniappan, Muthukannan Muthiah, Rajendran Gopinath

Environment Protection Engineering

|

2024

|

Vol. 50, nr 1

105--122

EN

Groundwater is a vital resource that provides drinking water to over half of the world's population. However, groundwater contamination has become a serious issue due to human activities such as industrialization, agriculture, and improper waste disposal. The impacts of groundwater contamination can be severe, including health risks, environmental damage, and economic losses. A list of unknown groundwater contamination sources has been developed for the Wang-Tien landfill using a groundwater modeling system (GMS). Further, AI-based models have been developed which accurately predict the contamination from the sources at this site. A serious complication with most previous studies using artificial neural networks (ANN) for contamination source identification has been the large size of the neural networks. We have designed the ANN models which use three different ways of presenting inputs that are categorized by hierarchical K-means clustering. Such an implementation reduces the overall complexity of the model along with high accuracy. The predictive capability of developed models was assessed using performance indices and compared with the ANN models. The results show that the hybrid model of hierarchical K-means clustering and ANN model (HCA-ANN) is a highly accurate model for identifying pollution sources in contaminated water.

8

DCSNN optimized with hybrid Border Collie optimization and Archimedes optimization algorithms for solid waste prediction in Chennai

Vidya Soundarapandian, Janani Elangeswaran Srie Vidhya, Sreedevi B.

Environment Protection Engineering

|

2024

|

Vol. 50, nr 1

5--25

EN

The rapid growth of smart cities and industry causes an increase in waste production. The amount of municipal solid waste (MSW) increases by several factors, including population growth, economic status, and consumption trends. The inadequacy of basic trash data is a major issue for managing MSW. Numerous existing models based on solid waste prediction have been presented so far, but none of them predict solid waste accurately and also it consumes more time. To address these concerns, a deep convolutional spiking neural network for solid waste prediction (DCSNN-SWP) is proposed in this paper. Here, the real-time solid waste prediction data are gathered from the quantity of municipal corporation of Chennai (MCC), landfill, garden garbage, and coconut shell reports in Tamil Nadu (Chennai), such as Zone 9 (Nungambakkam), Zone 10 (Kodambakkam) and Zone 13 (Adyar). Then the collected solid waste data are pre-processed using the kernel correlation model. Then the pre-processing data is given to DCSNN-hybrid BCMO and Archimedes optimization algorithm which accurately predicts the solid waste as wet waste, dry waste, horticulture waste, and dumping yard for 2022-2032 years. The proposed DCSNN-SWP method has been implemented in Python.

9

A method for determining the location and type of fault in transmission network using neural networks and power quality monitors

Šipoš Mario, Klaić Zvonimir, Nyarko Karlo Emmanuel, Fekete Krešimir

Eksploatacja i Niezawodność

|

2024

|

Vol. 26, no. 3

art. no. 187166

EN

In this paper, a procedure for determining the location of a fault on a power line using neural networks is proposed. Specifically, the procedure involves four stages (three of which employ neural networks): gathering voltage input data from power quality monitors via simulation, classifying the fault type, detecting the faulted line, and determining the fault position on the power line. The IEEE 39 bus test system was used to develop and test the mentioned model. Input voltages are obtained using DigSILENT PowerFactory software in which a set of three-phase and single-phase short circuits are simulated. For the next steps of the method, voltages from all buses are not used, but only voltages from optimally placed power quality monitors on 12 buses in the IEEE 39 bus test system. In the second step, the first neural network is employed in order to classify the fault type – single-phase or three-phase. In the third stage, another neural network is used to determine the faulted line and in the fourth stage, the last neural network is developed to determine the fault position on the faulted line.

10

Advancing Electrical Losses Assessment Methods in Power Systems

Rexhepi Vezir, Rexhepi Shaqir

Ecological Engineering & Environmental Technology

|

2024

|

Vol. 25, iss. 3

326--335

EN

The operation of modern power systems requires a sophisticated technological infrastructure to effectively manage and evaluate their parameters and performance. This infrastructure includes the generation, transmission and distribution power system components. This paper provides an overview of the loss evaluation to a part of Kosovo’s power system, substation with wind and photovoltaic (PV) energy sources integrated (SS Mramori, SS Kitka, and SS Kamenica) and the analysis of the loss assessment methods. One the assessment method in the research encompass simulated loss scenarios and their corresponding values in network components, employing the simulation based on the respective software tools. In current trends, power systems are visualized through the Supervisory Control and Data Acquisition (SCADA) platform. However, in Kosovo, although losses are integral to the SCADA system, they are represented as a overall value in the online mode, not encompassed depict losses per-components in real-time. This limitation hinders effective online power system optimization regarding the losses. As consequence, the purpose of this study is proposal a logical method developed through neural networks. The methodology incorporates various parameters, including as inputs variables; voltages, currents, active and reactive powers, and their computed values for extracting losses (X(x1, x2, ..., xn)). These parameters undergo systematic processing through hidden layers (Y(x1, x2, ..., xn)), leading to the classification of components within the power system. Finally, at the output stage (A(x1, x2, ..., xn)), an assessment is conducted based on the level of losses observed in the components of the power system. This implementation method promises significant benefits for transmission systems, impacting not only reducing losses, power quality but also yielding economic advantages.

11

Arduino-based implementation of kinematics for a 4 DOF robot manipulator using artificial neural network

Rashid Zaid Hikmat, Sarhan Riyadh Ahmed, Hassan Mohammed Salih

Diagnostyka

|

2024

|

Vol. 25, No. 1

art. no. 2024114

EN

Real-time motion control is basically dependent on robot kinematics analysis where there is no unique solution of the inverse kinematics of serial manipulators. The predictive artificial neural network is a powerful one for finding appropriate results based on training data. Therefore, an artificial neural network is proposed to implement on Arduino microcontroller for a 4-DOF robot manipulator where the inverse kinematics problem was partitioned to suit the low specification of the board CPU. With MATALB toolbox, multiple neural network configuration based were trained and experienced for the best fit of the dimensionless feedforward data that is obtained from the forward kinematics of reachable workspace with average MSE of 6.5e-7. The results showed the superior of the proposed networks reducing the joints error by 90 % at least with comparing to traditional one. An Arduino on-board program was developed to control the robot independly based on pre validated parameters of the network. The experimental results approved the proposed method to implement the robot in real time with maximum error of (± 0.15 mm) in end effector position. The presented approach can be applied to achieve a suitable solution of n-DOF self-depend robot implementation using micro stepping actuators.

12

Fault diagnosis of electrical faults of three-phasei nduction motors using acoustic analysis

Glowacz Adam, Sulowicz Maciej, Kozik Jarosław, Piech Krzysztof, Glowacz Witold, Li Zhixiong, Brumercik Frantisek, Gutten Miroslav, Korenciak Daniel, Kumar Anil, Lucas Guilherme Beraldi, Irfan Muhammad, Caesarendra Wahyu, Lui Hui

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2024

|

Vol. 72, nr 1

art. no. e148440

EN

Fault diagnosis techniques of electrical motors can prevent unplanned downtime and loss of money, production, and health. Various parts of the induction motor can be diagnosed: rotor, stator, rolling bearings, fan, insulation damage, and shaft. Acoustic analysis is non-invasive. Acoustic sensors are low-cost. Changes in the acoustic signal are often observed for faults in induction motors. In this paper, the authors present a fault diagnosis technique for three-phase induction motors (TPIM) using acoustic analysis. The authors analyzed acoustic signals for three conditions of the TPIM: healthy TPIM, TPIM with two broken bars, and TPIM with a faulty ring of the squirrel cage. Acoustic analysis was performed using fast Fourier transform (FFT), a new feature extraction method called MoD-7 (maxima of differences between the conditions), and deep neural networks: GoogLeNet, and ResNet-50. The results of the analysis of acoustic signals were equal to 100% for the three analyzed conditions. The proposed technique is excellent for acoustic signals. The described technique can be used for electric motor fault diagnosis applications.

13

Neural network applied to telescope pointing inaccuracy model

Zhaborovskyy Vitaliy, Medvedsky Myhailo, Choliy Vasyl, Pap Victor, Semenenko Viachelsav

Artificial Satellites : Journal of Planetary Geodesy

|

2024

|

Vol. 59, No. 2

55--63

EN

In the course of satellite observations using satellite laser ranging (SLR), a key task is pointing the telescope with high precision. Positioning the steering system’s mechanical parts with zero error is impossible. Accordingly, we must analyze and account for pointing errors by incorporating the telescope mounting errors themselves into the modeling error. Such models are far from trivial owing to the factors such as satellite azimuth, altitude, perhaps distance, or meteorological data. In this article, we explain how the data for the telescope pointing inaccuracy model (TIM) was collected and how a neural network was used to build a very precise TIM for the Golisiiv 1824 SLR station in Kyiv. We have focused our efforts on the suggested approach’s positive aspects based on our experience of using it to find practical solutions. Our practical recommendations may also be interesting for anyone working with hardware, especially in analyzing their errors. The key proof of the effectiveness of the approach is the serious increase in the number of satellites successfully tracked, especially for “blind” paths, when the satellite is not visible to the observer through the telescope guide.

14

Modelling the hot metal desulfurization process using artificial intelligence methods

Podolska Angelika, Falkus Jan

Archives of Civil Engineering

|

2024

|

Vol. 70, nr 2

255--270

EN

The objective of conducted research on the hot metal desulfurization process was to determine the key process parameters that impact the ultimate outcome of desulfurization. As a result, the noticeable outcome of implementing these measures should be the improvement of quality control. In order to determine these parameters, used artificial intelligence methods like as neural networks (ANN). On the basis of the production data collected from the actual metallurgical aggregate for hot metal desulfurization, neural networks were built that used quantitative data (mass of hot metal, mass of used reagents, etc.) and qualitative data (chemical analysis of hot metal). The parameters of the desulfurization process were divided into state parameters and control parameters. From the point of view of the technology of conducting the desulfurization process and building an on-line model, only control parameters can be changed during desulfurization. To describe the problem of predicting change in the sulfur content during the hot metal desulfurization process is sufficient an MLP type neural network with a single hidden layer. Adopting a more complex network structure would probably lead to a loss of the ability to generalise the problem. The research was carried out in STATISTICA Automated Neural Networks SANN.

PL

Budowa i weryfikacja modeli procesu odsiarczania surówki żelaza opartych o sztuczną inteligencję dostarcza bardzo wiele interesujących wyników, które mogą być wykorzystane zarówno w obszarze związanym ze sterowaniem jak i do celów teoretycznej analizy procesu. Wniosek dotyczący analizy teoretycznej w odniesieniu do modeli wykorzystujących sztuczne sieci neuronowe może wydawać się zbyt optymistyczny, bowiem powszechnie uważa się, że modele typu black box nie wzbogacają naszej wiedzy o procesie. W omawianym przypadku mamy jednak do czynienia z sytuacją, w której jednym z kluczowych pytań badawczych jest odpowiedź jak duży wpływ na proces mają czynniki statyczne, a jaki dynamiczne. Przy tak sformułowanym problemie, analiza warstwy wejściowej sieci neuronowej połączona z analizą istotności poszczególnych wielkości wejściowych, umożliwia co najmniej ocenę jakościową. W przeprowadzonych badaniach, dokładność modeli dedykowanych wymienionym w tablicy 3 grupom wytopów jest bardzo zadowalająca. Przewyższa ona dokładność dotychczasowych rozwiązań i może być podstawą do modernizacji istniejącego systemu sterowania. Średni błąd prognozy wyrażony za pomocą wartości błędu bezwzględnego to wynik rzędu 15 ppm dla wytopów o zawartości siarki <100 ppm oraz 25 ppm dla wytopów o siarce końcowej >100ppm. Wyniki te należy traktować jako dokładne ze względu na metodę określania składu chemicznego (Optyczna spektroskopia emisyjna OES) oraz jej zakres błędu pomiarowego.

15

Application of Deep Learning Methods for Trajectory Planning Based on Image Information

Babiarz Artur, Kustra Małgorzata, Wen Shuhuan

Advances in Science and Technology. Research Journal

|

2024

|

Vol. 18, no 6

247--258

EN

This work aims to develop a mobile robot utilizing neural network technology. The algorithm, programmed in Python on a Raspberry Pi 4B platform, is detailed across four main chapters. These chapters cover the fundamental assumptions of deep learning, the construction of the platform, and the research validating pattern recognition accuracy under various disturbances. The mobile platform employs a neural network to analyze selected traffic signs and translates the recognized patterns into corresponding motor movements.

16

Acoustic-Based Drone Detection Using Neural Networks – A Comprehensive Analysis

Paszkowski Waldemar, Gola Arkadiusz, Świć Antoni

Advances in Science and Technology. Research Journal

|

2024

|

Vol. 18, no 1

36--47

EN

The article presents and describes the implementation of research on the detection of a drone in an urban environment using of the sound features. The methods of drone detection were recognized on the basis of modeling and evaluation of the features of the audio and acoustic signal. The authors proposed the use of a neural network model for the needs of drone detection taking into account acoustic measurements in an anechoic chamber and in an urban environment. The final part presents the obtained results of the drone detection. For the purposes of detection, a neural network model was used in order to recognize the obtained images of the spectograms of sound sources.

17

Assessment of daily concentrations of air pollutants over four stations of Andhra Pradesh, India, during 2019-2022

Gogineni Rajesh, Umakanth Nandivada, Ramakrishna Yarlagadda, Veeraswamy Parise

Acta Geophysica

|

2024

|

Vol. 72, no. 2

1227--1249

EN

In recent years, pollution levels have increased to dangerous levels in several Indian cities. These levels are posing a severe threat to human’s health. Using the data from Central Pollution Control Board (CPCB), the current work focuses on highlighting the primary air pollutants in various regions such as Visakhapatnam (VSK), Hyderabad (HYD), Amaravati (AMV), and Tirupati (TPTY). Data from the Zoo Park area were used to study the location of HYD. Sulphur dioxide (SO2), carbon monoxide (CO), nitrogen dioxide (NO2), nitrogen oxides (NOX), particulate matter with particles less than 2.5 gm in diameter (PM2.5), particulate matter with particles less than 10 gm in diameter (PM10), and ozone (O3) are the air pollutants used for analysis in this work. An attempt was made to research the meteorological factors that contribute to the rising levels of air pollution between 2019 and 2022. Wind speed (WS), temperature (TEMP), relative humidity (RHUM), rainfall (RF), and solar radiation (SR) are the meteorological variables used in the analysis. The prediction of PM2.5 and PM10 was done using artificial neural network (NN) method. The NN method's outcomes show strong correlation in the forecasting of air pollution across four locations. The VSK station exhibited a high correlation of 91.29%, whereas TPTY station had a low correlation of 82.1%, based on CPCB PM2.5 observation and NN technique. The VSK station revealed a high correlation of 90.30%, whereas TPTY station had a low correlation of 81.61%, based on CPCB PM10 observation and NN technique.

18

Zastosowanie sieci neuronowych do detekcji impulsów o znanym kształcie w obecności silnego szumu i trendu

Betka Szymon, Blok Marek

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2023

|

nr 4

220--223

PL

Detekcja impulsów w odebranym sygnale radiowym, zwłaszcza w obecności silnego szumu oraz trendu, jest trudnym zadaniem. Artykuł przedstawia propozycje rozwiązań wykorzystujących sieci neuronowe do detekcji impulsów o znanym kształcie w obecności silnego szumu i trendu. Na potrzeby realizacji tego zadania zaproponowano dwie architektury. W pracy przedstawiono wyniki badań wpływu kształtu impulsu, mocy zakłóceń szumowych oraz trendu obecnego w sygnałach wejściowych sieci, na skuteczność detekcji zaproponowanych rozwiązań.

EN

Detecting pulses in a received radio signal, especially in the presence of strong noise and trend, is a difficult task. The article presents proposed solutions based on neural networks for the detection of pulses of known shape in the presence of strong noise and trend. Two architectures are proposed for the purpose. The paper presents the results of the study of the influence of the pulse shape, the noise power, and the trend present in the input signals of the network on the detection performance of the proposed solutions.

19

Wskazywanie punktu startowego dla iteracyjnych algorytmów obliczania położenia z użyciem sieci neuronowej

Sadowski Jarosław, Stefański Jacek

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2023

|

nr 4

244--247

PL

W referacie przedstawiono wyniki badań nad możliwością wskazywania punktu startowego do pierwszej iteracji dla algorytmu iteracyjnego obliczania położenia w systemie lokalizacji dwuwymiarowej. Do wskazywania punktu startowego użyto jednokierunkowej sieci neuronowej a celem badań było znalezienie jak najmniejszej struktury sieci, pozwalającej na zbieżność algorytmu estymacji położenia w całym obszarze badań.

EN

The paper presents the results of a study on the possibility of starting point selection for the first iteration for an iterative position calculation algorithm in a two dimensional location system. A feedforward neural network was used to indicate the starting point and the aim of the study was to find the smallest possible network structure, allowing the position estimation algorithm to converge over the entire study area.

20

Dekodowanie kodów iterowanych z użyciem sieci neuronowej

Fiedot Oskar, Blok Marek

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2023

|

nr 4

190--193

PL

Nadmiarowe kody iterowane są jedną z prostych metod pozyskiwania długich kodów korekcyjnych zapewniających dużą ochronę przed błędami. Jednocześnie, chociaż ich podstawowy iteracyjny dekoder jest prosty koncepcyjnie oraz łatwy w implementacji, to nie jest on rozwiązaniem optymalnym. Poszukując alternatywnych rozwiązań zaproponowano, przedstawioną w pracy, strukturę dekodera tego typu kodów wspomaganą przez sieci neuronowe. Zaproponowane rozwiązanie pozwala na wykrywanie oraz korekcję błędów w odbieranych ciągach.

EN

Redundant iterated codes are one of the simple methods of deriving long correction codes that provide high error protection. At the same time, although their basic iterative decoder is conceptually simple and easy to implement, it is not an optimal solution. Looking for alternative solutions, a neural network-assisted decoder structure for this type of codes was proposed. The solution presented in this paper allows the detection and correction of errors in the received sequences.