A Wearable Acoustic Sensor for Identification in Harsh Noisy Environments

This article proposes a wearable acoustic sensor for identification in harsh environments. Hexagonal structural design and ScAlN piezoelectric film are used to improve the sensitivity of the acoustic sensor. In addition, the proposed sensor also has the advantages of ultra-wideband (10 Hz-20 kHz), ultra-flat frequency response (±0.5 dB). Compared with traditional non-contact microphones, wearable acoustic sensors can clearly collect speech information in noisy environments. As a concept demonstration, we collected voice information from different people in a noisy environment and achieved 100% identity recognition with the assistance of machine learning. This sensor effectively solves the problem of noise interference on sound transmission in special working scenarios, and provides a new solution for wearable acoustic devices and human-computer interaction.