Contribution to Book
Characterization of Ear-Canal Feedback Pressure due to Umbo-Drive Forces: Finite-Element vs. Circuit ModelsARO Midwinter Meeting 2016 (2016)
Background: Hearing-aid users often complain of poor sound quality and difficulty understanding speech in noisy situations. One of the main reasons for this is that the microphone in a hearing aid is typically located above the pinna, rather than inside the ear canal, in order to minimize feedback. This, in turn, reduces the subject’s ability to perceive the acoustic pinna cues above about 4 kHz that are needed for sound localization. Various strategies for minimizing the feedback pressure (thereby increasing the Maximum Stable Gain, MSG) of a wide-bandwidth non-surgical Tympanic Lens are investigated numerically to facilitate placement of a microphone in the ear canal.
Methods: The ear-canal feedback pressure due to mechanical stimulation of the tympanic membrane (TM) was calculated using a 1D circuit model representing the TM as a distributed-parameter transmission line and the ossicular chain and cochlea as a network of lumped elements (O’Connor and Puria, 2008), and using a 3D finite-element model (FEM) representing the human middle ear (Puria et al., 2014). In both cases, the ear canal was terminated with an impedance boundary condition of rho*c for air. Circuit-model and FEM responses to umbo-force stimulation are compared with experimental data from four human cadaveric temporal bones (Puria et al., 2015).
Results: The MSG calculated from the two models and TB measurements decreases from about 40 dB at 0.1 kHz to about 15 dB at 1 kHz. Between 1.5 and 4 kHz the MSG is about 10 dB. As the frequency increases above 4 kHz, the MSG of the FEM, in agreement with experimental data, increases to 40 dB at 8 kHz. However, the MSG of the circuit model only reaches a maximum of 10 dB at 6 kHz. Furthermore, the direction of the force vector at the umbo can be optimized to increase the MSG by up to 10 dB in the 1–4 kHz range.
Conclusions: While the feedback pressure in devices that mechanically stimulate the umbo is significantly lower than in acoustic hearing aids, it is still not zero, and this limits the placement of a microphone in the ear canal for patients with significant hearing loss. Attempts at reducing the feedback pressure by altering the umbo-force vector have been studied using two computational models. While both of the models predicted the feedback pressure precisely below about 4 kHz, only the FEM results come close to the measurements at higher frequencies.
Publication DateSpring February 24, 2016
Citation InformationMorteza Khaleghi, Kevin N O’Connor and Sunil Puria. "Characterization of Ear-Canal Feedback Pressure due to Umbo-Drive Forces: Finite-Element vs. Circuit Models" San DiegoARO Midwinter Meeting 2016 (2016)
Available at: http://works.bepress.com/mkm/8/