Aleksandra Król-Nowak

Dr. Aleksandra Król-Nowak

orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 December 2017 in Archives of Acoustics
Verified
Journal Article
Open access
PDF file
PKPiotr Kleczkowski
Piotr Kleczkowski
AZAgnieszka Zak
Agnieszka Zak
Aleksandra Król-NowakAleksandra Król-Nowak
Aleksandra Król-Nowak
The first extensive investigation on the Lombard effect with Polish speech has been performed. Characteristic parameters of Lombard speech were measured: intensity, fundamental frequency, spectral tilt, duration of words, duration of pauses and duration of vowels. The effect was investigated in a task involving real communication - solving a Sudoku puzzle. The speakers produced speech in quiet and in three different backgrounds: competing speech, speech-shaped noise and speech-modulated noise. The experimental conditions were held as close as possible to those in the study by Cooke and Lu (2010) so that conclusions could be drawn whether differences between the Lombard effect in Polish speech and English speech existed. Most of the findings on the Lombard effect known from the literature have been confirmed with Polish speech. In three parameters, Polish speakers were more sensitive to modulated backgrounds while English speakers were more sensitive to a stationary background. In both languages, the modulated backgrounds induced speakers to extend pauses in the communication tasks.

16 July 2025 in Sensors
Verified
Journal Article
Open access
PDF file
Piotr KsiążekPiotr Książek
Piotr Książek
Urszula LibalUrszula Libal
Urszula Libal
Aleksandra Król-NowakAleksandra Król-Nowak
Aleksandra Król-Nowak
This study investigates the impact of microphone placement on honey bee audio monitoring for time-of-day classification, a key step toward automated activity monitoring and anomaly detection. Recognizing the time-dependent nature of bee behavior, we aimed to establish a baseline diurnal pattern recognition method. A custom apparatus enabled simultaneous audio acquisition from internal (brood frame, protected from propolization) and external hive locations. Sound signals were preprocessed using Power Spectral Density (PSD). Extra Trees and Convolutional Neural Network (CNN) classifiers were trained to identify diurnal activity patterns. Analysis focused on feature importance, particularly spectral characteristics. Interestingly, Extra Trees performance varied significantly. While achieving near-perfect accuracy (98–99%) with internal recordings, its accuracy was considerably lower (61–72%) with external recordings, even lower than CNNs trained on the same data (76–87%). Further investigation using Extra Trees and feature selection methods using Mean Decrease Impurity (MDI) and Recursive Feature Elimination with Cross-Validation (RFECV) revealed the importance of the 100–600 Hz band, with peaks around 100 Hz and 300 Hz. These findings inform future monitoring setups, suggesting potential for reduced sampling frequencies and underlining the need for monitoring of sound inside the beehive in order to validate methods being tested.

24 December 2025 in Sensors
Verified
Journal Article
Open access
PDF file
PKPiotr Kleczkowski
Piotr Kleczkowski
Aleksandra Król-NowakAleksandra Król-Nowak
Aleksandra Król-Nowak
Air pollution is monitored worldwide through networks of sensors. They provide information on local air pollution, which also provides a basis for a multitude of research. To reduce health hazards caused by air pollution, the concentrations of pollutants as measured by sensors need to be apportioned to particular sources. Although several methods to achieve this have been developed, only a few works on the contributions of pollution sources to health hazards are available in the literature. In this work, a simple scheme is proposed to compare health hazards from each of the main sources of air pollution in a given country, region, or area. The comparison involves the main air pollutants of PM2.5, NO2, and O3 for chronic exposures and PM2.5, NO2, O3, and SO2 for acute exposures. The actual health hazard from each substance is determined from concentrations measured by sensors and the concentration–response functions found in the literature. The apportionment of substances to sources is based on emission inventories, thus avoiding costly methods of source apportionment. This method has been applied to the entire country, i.e., Poland, yielding the average proportion of health hazards from particular sources. The example demonstrates the flexibility and ease of application of the scheme. Uncertainties in the results were subjected to discussion. The key advantage of the scheme lies in its ability to provide an indication of the most harmful sources of pollution, thus highlighting efficient interventions.