Inspired by the human cochlea, we propose a directional sound sensor using a resonator array to overcome the limitations of existing directional microphones. The proposed sensor consists of multiple cantilevers that respond to different resonance frequencies and separately acquire signals to then combine them for sound sensing. The directionality of the cantilevers is bipolar (figure-of-8) because a signal proportional to the input sound’s pressure gradient is generated. We adopt multimode resonance to cover the wide frequency range of 100-8,000 Hz using few resonators. A wide-bandwidth (low-quality-factor) trenched cantilever is used to obtain a flat frequency response. Bidirectional sound sensors are tapered to achieve acoustic beamforming by simple signal processing. The directional characteristics can be easily changed according to the weighted sum of the signals acquired from a pair of sensors. We demonstrate that ambient noise can be effectively suppressed through beamforming to acquire the desired signal using the proposed sensor. [2021-0012]

中文翻译：

---

在可听范围内具有一致方向性和灵敏度的定向声音传感器

受人类耳蜗的启发，我们提出了一种使用谐振器阵列的定向声音传感器，以克服现有定向麦克风的局限性。所提出的传感器由多个悬臂梁组成，这些悬臂梁响应不同的共振频率并分别获取信号，然后将它们组合起来进行声音传感。悬臂的方向性是双极的（8 字形），因为生成的信号与输入声音的压力梯度成正比。我们采用多模谐振以使用很少的谐振器覆盖 100-8,000 Hz 的宽频率范围。宽带（低品质因数）沟槽悬臂用于获得平坦的频率响应。双向声音传感器逐渐变细，通过简单的信号处理实现声波束形成。根据从一对传感器获取的信号的加权总和，可以轻松改变方向特性。我们证明可以通过波束成形有效抑制环境噪声，从而使用所提出的传感器获取所需信号。[2021-0012]