Directivity Index (DI) and Polar Measures
The Directivity Index (DI) of a directional microphone array is the ratio of the microphone output for a sound source positioned directly in front of the microphone array to that array’s output measured in a diffuse sound field of the same total acoustic power. A diffuse sound field is one in which the statistical distribution of sound energy is uniform throughout the sound field and the directions of wave propagation are randomly distributed. While this definition of a diffuse field DI is a theoretical ideal, a representative measure can be derived in a free-field using an anechoic chamber and single loudspeaker. The methodology for this approach is outlined below. Knowledge of a directional microphone’s DI is useful in that this information may be interpreted as an estimate of the effective signal to ambient noise ratio improvement when the directional microphone is active. It is reasonable to expect that in many noisy listening situations increasing a directional microphone’s DI will improve signal-to-noise ratio and result in improved speech recognition ability and comfort for the listener.
Methodology for Directivity Index (DI) Measures
The following methodology details the data collection procedures for quantification of directivity measures. These methods apply to all test devices whenever possible. If deviation from these methods is required for data collection, it will be explicitly noted in the benchmarking report.
Free Field DI and Polar Measures
All testing is completed in a 8'9" x 8'9" x 6'9" anechoic chamber. A measure of DI requires that the hearing aid is rotated in either a full circle or a half circle (with the other half mirror imaged due to symmetry) in an Azimuth plane with the microphone output being recorded at every 5 degrees of rotation. Each hearing aid is attached to the end of a mechanical arm at the battery compartment of the hearing aid and is coupled to a 2cc coupler using soft putty to ensure a reliable seal between the device and coupler opening. The distance from the loudspeaker to the hearing aid reference point is 1m. This reference location is halfway between the directional microphone ports, which lies in the rotational axis as well. This ensures that the rotation of the hearing aid will maintain the specified distance between each of the two microphone ports and the sound source. A B&K 3560C Pulse analyzer is used to generate a Maximum Length Sequence (MLS) stimulus via a KEF point source loudspeaker. Signal presentation will vary between 50 and 75 dBSPL depending on device type, style and fitting configuration. A Visual Basic program is used to automate the data collection process; initiating a recording at every 5 degree increment for a total of 360 degrees (A 3-D recording will include complete rotations in both the horizontal and vertical axis). The test signal is recorded using a 0.5” B&K condenser microphone and routed to the B&K Pulse analyzer. The recorded signal is then processed by Matlab to generate information related to measurement condition and directional performance including DI and polar diagrams.
Below is an example of how the data from free-field directivity measures may be used:
Table 1. Average Free-field Directivity Index (DI) for the Zōn and the three closest performing competitors.
The following is an excerpt from: Galster, J.A., Yanz, J.L., and Freeman, B.A. In the Zōn: Excellence and innovation in hearing aid design. Technology Update, 2008.
Directivity in the free field. Table 1 provides horizontal plane, free-field DIs, measured in a sealed coupler, the data are averaged across 500, 1000, 2000, and 4000 Hz, for Zōn and the top three performing competing products. DI is a valuable measure in that it provides a single number that is representative of the attenuation provided by directional microphones. This attenuation may also be interpreted as effective SNR. In other words, the magnitude of DI is related to the amount of directional benefit a patient may experience (Ricketts, 2001).
» Read the full article (551KB PDF)
Knowles Electronic Manikin for Acoustics Research (KEMAR) Measures
The test environment, hardware and output recording methods for insitu measurement of directivity are similar to those used for free-field measurement. Each hearing aid is placed in a clinically appropriate position on the right pinna of the KEMAR, instead of a 2cc coupler. It is then coupled to a Zwislocki coupler between the device receiver or ear hook and earmold or coupler opening (soft putty maybe used to ensure a reliable seal). All signal recording is completed using a 0.5" condenser microphone and routed to a signal analyzer for processing. When collecting 3-D data, the speaker is moved from +90 the KEMAR is rotated a full 360 degrees in both the vertical and horizontal planes.
Below is an example of how the data from directivity measures recording using a KEMAR may be used:
Figure 1. Directivity Index as a function of frequency for the Starkey Zōn and three competing RIC products.
The following is an excerpt from: Galster, J.A., Yanz, J.L., and Freeman, B.A. In the Zōn: Excellence and innovation in hearing aid design. Technology Update, 2008.
Directivity on KEMAR. Figure 1 shows DI as a function of frequency, recorded on a KEMAR manikin in an open-canal fitting with Starkey Zōn and the three top performing competing products. As expected, the DI quickly falls off below 1000 Hz for all products. A directional response cannot be obtained in a frequency region where there is no gain (Bentler, 2006). Thus, the lack of low-frequency directivity is the result of the basic open-canal design, the acoustics of which inherently roll off low-frequency gain. Throughout the key speech frequency range the directivity of all the evaluated instruments grows, with Zōn consistently achieving the highest DI versus its competition.
» Read the full article (551KB PDF)