Matematikcentrum

Lund University

Examensarbetesförslag

  • Kors-spektrumanalys av hjärtfrekvens i samband med arbetsrelaterad stress
  • A number of recent studies report that decreased heart rate variability (HRV) power is related to cardivascular disease, depression, various anxiety disorders, and long-term work related stress or burnout. Cross spectral analysis between the HRV and the breathing signals can be used as a refined technique for analyzing the HRV power. The project aim is to compare and develop such methods for the analysis of HRV data collected from individuals at different stages of work related burnout. This project aims for classification of groups of patients with stress related diagnosis using a novel methodology for time-frequency analysis of locally stationary processes. The work includes analysis and evaluation on a novel set of HRV measurement data controlled by metronome guided respitation. Prerequisites: FMSF10/MASC04, FMSN35/MASM26
  • Songbird dialects
  • Many songbird species have geographically structured vocal culture in which individuals at a location produce songs similar to each other that clearly differ from songs produced at other locations by members of the same species.  Scientific study of such vocal culture is facilitated by reliable quantification of the similarity between pairs of songs. However, quantification of similarity between complex songs recorded in noisy environments in the wild is a substantial challenge. For instance, in the North American bird Spiza americana, human observers can still readily outperform existing quantitative methods in assessing the similarity between songs. Therefore, the goal of this project is to improve existing quantitative methods for assessing similarity between the songs of Spiza americana. The project is a collaboration with Timothy Parker, Dept of Biology, Whitham College, Walla Walla, USA Prerequisites: FMSF10/MASC04, (FMSN35/MASM26)
  • Respiratory and pulse monitoring
  • The goal of this project is to achieve respiratory and/or pulse monitoring using novel radar technology. Challenges lie both in extracting the displacement of the chest and abdomen from the raw radar signal, and then also in extracting respiratory/pulse information from that data. General knowledge of methods for analysis of stochastic processes as well as understanding of filtering techniques are essential. The project will be done at Acconeer using their 60 GHz pulsed coherent radar which originates from research at LTH. Prerequisites: FMSF10/MASC04, (FMSN45/MASM17, FMSN35/MASM26)
  • Multiple-channel dolphin sonar beam characterization
  • Transient signals are by nature difficult to characterize. The sonar beam of toothed whales contains several signal components and to accurately detect and localize the components in the time-frequency domain is essential to understand to what extent the signal can be controlled by the animal and what functions it serves. The scaled reassigned spectrogram estimates the time- and frequency centers of individual signals and is therefore well suited for detection and localization of transient signal components, also when they are closely located in the time-frequeny domain. This project aims to studying and characterizing multiple-channel sonar beam measurements and also possibly develop and tailor the method using information from the multiple-channel structure. Prerequisites: FMSF10/MASC04, (FMSN35/MASM26)