关于「堵耳效应:闭塞耳道的集总参数方程与开孔的代价」,《耳内降噪、堵耳效应与听力努力》给出的定量结论是什么?

结论:第四章用集总参数模型计算堵耳效应,泡棉泄漏下浅插在 250 Hz 为 13.0 dB,并说明开孔缓解闷感的频段正是放弃低频隔声的频段。

数据口径:书中的耳道尺寸、噪声谱、环路增益与认知参数均为示意;全部计算由书中程序生成;本书不是医疗建议,不宣称任何产品或聆听方式对听力、耳鸣或健康有任何益处。 第一章在合成噪声上运行最小均方与滤波 x 最小均方算法,说明教科书步长界 2/(L·P) 对高斯输入偏松:16 抽头时精确界是 0.1111,而次级通路模型的相位误差越过 90 度时算法发散。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 13.0, 250

source U: open 11.5, shallow 7.5, deep 1.00
occlusion effect in dB6312525050010002000
shallow 4 mm, sealed59.647.735.623.29.8-13.1
shallow 4 mm, foam leak24.318.513.08.87.8-7.0
deep 15 mm, sealed45.433.521.49.0-4.4-27.2
deep 15 mm, foam leak6.81.0-4.5-8.9-11.0-17.6
sealed shallow: open-mass x residual-compliance f0 = 2414 Hz
sealed shallow slope 125 to 250 Hz: 12.12 dB per octave

核心公式

\[\mathrm{OE}(f)\approx 20\log_{10}\frac{U_{\mathrm{occ}}}{U_{\mathrm{open}}}\cdot\frac{1}{\omega^{2}MC}.\]
查看完整验算程序
# Chapter 4. p41: Occlusion effect from a lumped ear-canal model.
# Model (all values illustrative, of textbook order): air 1.204 kg/m^3,
# c = 343 m/s, canal 25 mm long, 7.5 mm wide; outer 10 mm cartilaginous,
# inner 15 mm bony; eardrum load = 0.6 cm^3 of equivalent volume. Bone-
# conducted own voice drives the canal walls as a volume-velocity source
# U, 1 per mm of exposed cartilage and 0.1 per mm of exposed bone.
# Open ear: U sees the canal air mass to the outside, M = rho l / A
# (half the canal plus an end correction 0.61 r), in parallel with the
# compliance C = V / (rho c^2). Occluded: U sees the residual volume,
# in parallel with a foam-like leak, R = 4e6 Pa s/m^3 in series with a
# mass of 1e3 kg/m^4. OE(f) = 20 log10 |U_occ Z_occ| / |U_open Z_open|.
# Sealed and lossless, OE ~ 20 log10 (U_occ / U_open / (w^2 M C)), which
# falls 40 dB a decade; the leak flattens it at low frequency.
import math

RHO, C0 = 1.204, 343.0
LC, LCART, R = 25e-3, 10e-3, 3.75e-3
AREA = math.pi * R * R
V_ED = 0.6e-6
S_CART, S_BONE = 1.0, 0.1
R_LEAK, M_LEAK = 4e6, 1e3


def source(depth):
    cart = max(LCART - depth, 0.0)
    bone = LC - max(depth, LCART)
    return (S_CART * cart + S_BONE * bone) * 1e3


def par(z1, z2):
    return z1 * z2 / (z1 + z2)


def z_open(f):
    w = 2 * math.pi * f
    m = RHO * (LC / 2 + 0.61 * R) / AREA
    cpl = (AREA * LC + V_ED) / (RHO * C0 * C0)
    return par(1j * w * m, 1 / (1j * w * cpl))


def z_occ(f, depth, leak):
    w = 2 * math.pi * f
    cpl = (AREA * (LC - depth) + V_ED) / (RHO * C0 * C0)
    zc = 1 / (1j * w * cpl)
    if not leak:
        return zc
    return par(zc, R_LEAK + 1j * w * M_LEAK)


def oe(f, depth, leak):
    num = source(depth) * abs(z_occ(f, depth, leak))
    return 20 * math.log10(num / (source(0.0) * abs(z_open(f))))


if __name__ == "__main__":
    cases = (("shallow 4 mm, sealed", 4e-3, False),
             ("shallow 4 mm, foam leak", 4e-3, True),
             ("deep 15 mm, sealed", 15e-3, False),
             ("deep 15 mm, foam leak", 15e-3, True))
    freqs = (63, 125, 250, 500, 1000, 2000)
    print("source U: open {:.1f}, shallow {:.1f}, deep {:.2f}".format(
        source(0.0), source(4e-3), source(15e-3)))
    print("occlusion effect in dB" + " " * 9 + "".join(
        "{:>7}".format(f) for f in freqs))
    for name, dep, lr in cases:
        print("{:<30}".format(name) + "".join(
            "{:>7.1f}".format(oe(f, dep, lr)) for f in freqs))
    m = RHO * (LC / 2 + 0.61 * R) / AREA
    cpl = (AREA * (LC - 4e-3) + V_ED) / (RHO * C0 * C0)
    f0 = 1 / (2 * math.pi * math.sqrt(m * cpl))
    print("sealed shallow: open-mass x residual-compliance f0 = "
          "{:.0f} Hz".format(f0))
    print("sealed shallow slope 125 to 250 Hz: {:.2f} dB per octave".format(
        oe(125, 4e-3, False) - oe(250, 4e-3, False)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《耳内主动降噪最小均方自适应滤波、堵耳效应低频骨导增益与信噪比认知恢复》第 4 章「堵耳效应:闭塞耳道的集总参数方程与开孔的代价」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-164 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026164,
  author    = {Fan, Yafu},
  title     = {LMS Adaptive Filtering in In-Ear Active Noise Cancellation, Occlusion-Effect Bone-Conduction Gain and SNR-Driven Cognitive Recovery},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论