以quasi-linear法建立頻域內耳聽覺生理模型

Establishing an auditory model of the inner ear in the frequency domain by quasi-linear methods

陳永境(Yong-Zing Chen)

指導教授 ： 劉奕汶

國立清華大學/電機資訊學院/電機工程學系/碩士(2013年)

https://doi.org/10.6843/NTHU.2013.00781

摘要

過去藉由電腦模擬人類的聽覺機制往往使用時域上的模型，而利用頻域模型來進行模擬的方式則較少被使用與討論。時域模型因為針對每個時間點解微分方程組，因此擁有擬真度較高的優點，但相對而言運算量複雜度也比較大；反之，頻域模型則可藉由轉移函數快速得到模擬結果。但由於耳蝸為非線性系統，其放大機制無法以簡單的線性關係描述，因此會產生較大誤差。
本論文以Liu and Neely在2009年提出的內耳模型為基礎，利用數學技巧:quasi-linear method的迭代過程找出耳蝸的等效電流敏感度並改造上述模型，如此便可建構符合耳蝸非線性的頻域模型，更有效率也更貼近真實生理特性的模擬人類聽覺現象。
以此改造過的頻域模型模擬不同耳蝸特性包含：網狀膜(Reticular Lamina)位移及相位、電流敏感度常數(sensitivity ratio)、單位距離穿越波數(wave number)，並與Liu and Neely在2010年提出的中耳至內耳時域模型模擬結果進行比較，發現大部分模擬結果能與時域模型吻合，並且也更快速的模擬出結果。

關鍵字

耳蝸 ； 時域模型 ； 頻域模型 ； 電流敏感度常數 ； 外毛細胞

並列摘要

While the frequency domain equivalent models were seldom utilized and discussed, previous computer models to simulate human auditory mechanism were mainly constructed in the time domain. With a higher computational complexity, the time domain models usually match the real mechanism well, for they require solving a set of partial differential equations for each time step. The frequency domain models, on the other hand, offer faster simulation results through transfer functions. However, due to the nonlinearity of the cochlea, the amplification mechanism cannot be described with simple linear relationships, and the results from the frequency domain models could deviate from the real mechanism.
Based on the frequency model of the inner ear (Liu and Neely, 2009), this study improves the existing model by introducing the effective sensitivity of the cochlea, which could be approximated by the quasi-linear method. The frequency domain model with nonlinearity is established through this approach, providing more precise simulation results with high efficiency.
The improved frequency model is capable of simulating different cochlear characteristics, including the displacement and phase of Reticular Lamina, the sensitivity ratio, and the wave number per length. Comparing with the simulation results from the previous time domain model(Liu and Neely, 2010), the improved frequency domain model in this study provides compatible results and a efficient solution to the simulation of human auditory mechanism.

並列關鍵字

cochlea ； quasi-linear ； FDM ； TDM ； sensitivity ratio ； outter hair cells

參考文獻

[1] Liu, Y. W., and Neely, S. T. (2010). “Distortion product emissions from a cochlear model with nonlinear mechanoelectrical transduction in outer hair cells,” J. Acoust. Soc. Am., 127(4), 2420–2432.

[2] Liu, Y. W., and Neely, S. T. (2009). “Outer hair cell electromechanical properties in a nonlinear piezoelectric model,” J. Acoust. Soc. Am., 126(2), 751–761.

[3] Kanis, L. J., and de Boer, E. (1993). “Self‐suppression in a locally active nonlinear model of the cochlea: A quasilinear approach,” J. Acoust. Soc. Am., 94(6), 3199–3206.

[4] Meaud, J., and Grosh, K. (2012). “Response to a pure tone in a nonlinear mechanical-electrical-acoustical model of the cochlea,” Biophys J, 102(6), 1237–1246.

[6] Guinan, J. J. (2007). “Olivocochlear efferents: Anatomy, physiology, function, and the measurement of efferent effects in humans,” Ear and Hearing, 27(1), 589.