Page 26 - January 2006

atmospheric turbulence into waveforms provided by industry and NASA. This task will then feed in to the perception and sub- jective testing part of the work which includes two tasks, 8.6 and 8.7. Task 8.6 is entitled Determine Annoyance of Low Boom Waveforms, and Task 8.7 is entitled Noise Metrics.
The primary goal of Task 8.1 during
the first year will be to generate waveforms
of a shaped boom design that have been
propagated through realistic atmospheric effects. Task 8.1 is being carried out at The Pennsylvania State University, with the present author as Principal Investigator.
The effort has begun by employing the large database of recorded waveforms from the shaped sonic boom demon- strator (SSBD) aircraft flights of August 2003 and January 2004. Those flights were for a modified F-5E aircraft with flat-top front shock shaping only. The waveforms collected in those studies have been carefully analyzed in collabora- tion with industrial partners. The approach has been to develop filter functions for different realizations of atmos- pheric turbulence, and this has been achieved for first pass filter functions. The work is now at the point where one can convolve “clean” sonic boom waveforms with the inverse Fourier transform of the filter function to “turbulize” them. Future work will use a more physics-based model by exploit- ing existing models for propagation through atmospheric turbulence.
Subjective studies in PARTNER
Both The Pennsylvania State University and Purdue University are working on both Tasks 8.6 and 8.7. It is assumed that supersonic aircraft signatures outdoors will be studied first, as this is much easier than indoor signatures. Indoor signatures will be discussed below.
Shaped sonic boom waveforms have not undergone extensive subjective testing to the extent that traditional N waves have. The purposes of Task 8.6 are to more fully understand the human response to realistic low-boom sonic boom waveforms and to compare listener reactions between shaped sonic booms and other natural sounds such as dis- tant thunder.
Since no low-boom aircraft exist at this time, all subjec- tive testing at the outset of Project 8 will take place in sonic boom simulators. Three simulators are currently involved in the research: (1) a simulator developed by and residing at NASA Langley Research Center, Hampton, VA, (2) a simula- tor similar in construction to the Langley simulator at Lockheed-Martin Aeronautics, Palmdale, CA, and (3) a portable simulator built by Gulfstream Aerospace Corp. described by Salamone at the July 2005 ISBF.
In Task 8.7 the applicability of both traditional and very recently developed noise metrics to shaped sonic boom sounds will be determined. The repeatability of human sub- ject testing results across the three different sonic boom sim- ulators will also be assessed.
There are many metrics one could try to apply to sonic boom signatures. However, some have already been deter-
"...sonic boom is made continuously during the entire time that the aircraft is moving supersonically."
mined to be more successful than others in predicting annoyance. A leading con- tender, showing the best correlation to annoyance is the Stevens’ Mark VII
16
Perceived Loudness (PLdB) . Any new
metrics will have to be compared to this baseline metric which seems to be quite robust in terms of its applicability across a wide range of boom shapes and ampli- tudes.
PLdB does not account for many things one would want in an appropriate low-boom metric. It does not account for startle, whole-body response, or a nighttime penalty. Clearly, additional work needs to be
done.
Subjective studies underway
The initial work in tasks 8.6 and 8.7 is centered on show- ing the realism of the sonic boom simulators and on the mutual reproducibility between those simulators. Brenda Sullivan of NASA Langley, Kathleen Hodgdon of Penn State, and Patricia Davies of Purdue have been working during 2005 on the planning and execution of three human subject experiments.
Sullivan has led the effort to compare the NASA Langley simulator with the portable simulator of Gulfstream. Comparison tests by “expert ears” from the East Coast of the United States took place in May and early June 2005. These experts are individuals who have heard many real sonic booms previously, and they were asked to rate the realism of the playback of recorded sonic boom waveforms in both sim- ulators.
A second, similar comparison test between the Gulfstream simulator and the simulator at Lockheed-Martin in Palmdale, CA is being led by Davies using experts from the U.S. West Coast. That test occurred in late August and early September 2005. Combined and coordinated with the earlier work led by Sullivan, it should be possible to have a three-way comparison between the simulators regarding realism and mutual reproducibility.
A third test, conceived by Peter Coen of NASA Langley, was designed and executed by Hodgdon to compare real sonic boom noise by military supersonic overflights with playback reproduction in the Gulfstream sonic boom simulator. F-18 aircraft were flown supersonically seventeen times over a group of subjects for them to become familiar with real sonic boom noise on the morning of June 15, 2005 at NASA Dryden Flight Research Center (see Fig. 7). Simultaneous 24-bit recordings were made which were played back for those sub- jects later that afternoon in the Gulfstream simulator. Results from this and the other subjective experiments are expected sometime within the next year.
During the June 15, 2005 experiment, measurements of the seventeen supersonic flights were also made with new, high-bandwidth recording equipment. Further, acoustic measurements were made inside and immediately outside a nearby house on-site at Edwards Air Force Base. Vibration measurements were also carefully recorded inside the house. Although there are no plans to analyze this data right
24 Acoustics Today, January 2006