Spectral Analysis and Sonification of Simulation Data Generated in a Frequency Domain Experiment

In this thesis, we evaluate the frequency domain approach for data farming and assess the possibility of analyzing complex data sets using data sonification. Data farming applies agent-based models and simulation, computing power, and data analysis and visualization technologies to help answer complex questions in military operations. Sonification is the use of data to generate sound for analysis. We apply a frequency domain experiment (FDE) to a combat simulation and analyze the output data set using spectral analysis. We compare the results from our FDE with those obtained using another experimental design on the same combat scenario. Our results confirm and complement the earlier findings. We then develop an auditory display that uses data sonification to represent the simulation output data set with sound. We consider the simulation results from the FDE as a waveshaping function and generate sounds using sonification software. We characterize the sonified data by their noise, signal, and volume. Qualitatively, the sonified data match the corresponding spectra from the FDE. Therefore, we demonstrate the feasibility of representing simulation data from the FDE with our sonification. Finally, we offer suggestions for future development of a multimodal display that can be used for analyzing complex data sets.