Auralizing Impulsive Sounds Outdoors Among Buildings
Open Access
- Author:
- Lind, Amanda Blair
- Graduate Program:
- Acoustics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 10, 2017
- Committee Members:
- Victor Ward Sparrow, Dissertation Advisor/Co-Advisor
- Keywords:
- Sonic Boom
Auralization
BTM
Biot Tolstoy Mediwn
Diffraction
Edge Diffraction
Impulsive
Low Frequency
Outdoor
Sound Propagation - Abstract:
- Industry partners have identified a market for at least 450 supersonic business jets. Since The Concorde, current legislation prohibits overland supersonic flight, due to the human and environmental impact of the sonic booms generated all along the supersonic flight path. In the 1970s, new theory relating the cross-sectional area of the aircraft with the sonic boom waveform on the ground was introduced - allowing for next generation supersonic aircraft to be designed with sonic boom mitigation in mind. As such, NASA, the FAA, and industry partners such as Gulfstream, Lockheed Martin, and Cessna have partnered to quantify the human impact of these proposed designs prior to aircraft construction. The human impact of these next generation sonic booms are predicted through simulation of the Computational Fluid Dynamics (CFD) around the aircraft, nonlinear propagation, and propagation through the turbulent boundary layer prior to reaching listeners over infinite level ground. Simulations, and recordings of measured sonic booms are reproduced in sonic boom simulators for subjective testing. The role of this work is to offer a tool for simulating listenable sonic boom waveforms, a process known as auralization, in more realistic listening environments than infinite level ground. In order to increase the fidelity and perceived realism of synthesized sonic booms, a model superimposing direct sound, specular reflections and diffracted contributions was implemented. Simulations of sonic booms in various listening environments were performed with this model, and compared against recorded sonic booms. The impact of specular reflections and diffracted contributions on human impact iii of sonic booms was quantified with the industry standard PLdB metric. Finite impulse response (FIR) filters characterizing the listening environment and source/receiver orientation are generated using the image source method (ISM) and a time domain edge diffraction model by Biot, Tolstoy, and Medwin (BTM). Using the software tool provided in this work, the generation of an FIR filter predicting specular reflections may be calculated in any planar geometry: reflections from more complicated terrain and structures could be approximated by thoughtfully created planar geometries. The ray based Image Source Method (ISM), common to architectural acoustics, was adapted for outdoor applications and sonic boom excitations. Accounting for reflections from the ground and vertical structures was shown to dramatically alter subjective loudness metrics. This work was motivated by two goals: 1. Provide a tool to enable better prediction the impact of specular reflections on PLdB in more complicated geometries. The tool offers a means of improving the fidelity of, and expanding the collection of sonic booms available for subjective listening tests. 2. Identify if and when simulating diffracted contributions is required for accurate PLdB prediction. More complicated geometries simulated in this body of work have demonstrated that idealized specular reflections alone yield a variation in PLdB from -99.3 dB due to complete occlusion, to +14.2 dB due to constructive interference and coincident building reflections. These simulations were performed at listener height around a building with an 'L-shaped’ foot print, and at ground level around a multi-family residence. Specular fields simulated around an isolated wall showed excellent for microphone positions that were not occluded from the direct sound. Microphone positions behind occlusions were better modeled through an edge diffraction model. Analysis of the diffracted impulse responses (IRs) around the isolated wall showed that for receiver positions located closer to the diffracting edges and closer to shadow boundaries exhibited more impulsive diffracted IRs. PLdB is proportional to, and highly dependent on the rise time of the shocks of the sonic boom. The rise iv time of the boom is decreased when convolved with a less impulsive IR. As such, the diffracted contributions exhibit the high PLdB of the incident wavefront when close to the edge, and/or close to the shadow-zone boundary. Our edge diffraction model greatly improved prediction of sonic boom wave forms and metrics in the shadow zone, exhibiting an error of 14 PLdB. This compares to an error if 93 PLdB when the diffracted model is omitted.