An artificial acoustic sensor capable of selectively detecting danger signals has been developed.
On the 24th, the Korea Research Foundation announced that Professor Han Changsoo and Dr. Jeon Eunseok's research team at Korea University developed a next-generation acoustic sensor with a ‘self-powered, multi-channel’ method that can separate and detect frequencies by mimicking the human cochlea (sound perception process).
Professor Han Chang-su and Dr. Jeon Eun-seok, Korea University. Provided by the National Research Foundation of Korea.
The cochlea is located deep inside the ear and functions to convert sound vibrations (frequencies) into electrical signals that are transmitted to the brain.
When the spiral-shaped cochlea is uncoiled, there is a very thin cellular membrane called the basilar membrane along the internal canal. The base, which is the starting part, is wide and thin, but as it goes toward the apex of the cochlea, the width narrows and the thickness increases. This shape of the cochlear basilar membrane allows humans to detect various sounds across different frequency bands.
Research on acoustic sensors mimicking the biological function of the cochlea has continued for over 20 years. However, acoustic sensors developed in previous studies had narrow frequency bands and insufficient separation between frequency bands across multiple channels, limiting their ability to detect and analyze sounds.
Based on this, the research team succeeded in developing a next-generation artificial basilar membrane sensor that more precisely mimics the shape of the cochlear basilar membrane. This was achieved by developing and applying a new method that effectively reflects the three-dimensional structural characteristics of the biological basilar membrane in the design.
First, the artificial basilar membrane sensor developed by the team was designed to vary in width along its length, similar to the biological basilar membrane.
Additionally, by adopting a spiral structure, the length was maximized relative to the area, thereby expanding the frequency band. The frequency band of the existing artificial basilar membrane sensors, which was less than 1 decade, was extended to 2.13 decades, ranging from 96 to 12,821 hertz (Hz).
Another notable point is that 24 piezoelectric sensor modules were attached to mimic the basilar membrane and auditory nerves, allowing each of the 24 channels to have independent frequency bands.
This enables the formation of desired characteristic frequencies (frequencies of sounds with maximum amplitude depending on the vibrating position of the basilar membrane) according to the position on the basilar membrane, implying the introduction of the concept of minimum distance on the basilar membrane.
The research team also succeeded in identifying the driving sounds of high-speed, heavy vehicles such as buses, trucks, and motorcycles running on roads using the developed artificial acoustic sensor, distinguishing the types of vehicles. This verified the frequency separation capability and electrical signal detection ability.
Professor Han Changsoo said, “The artificial acoustic sensor developed by the research team is expected to be used as an early warning system that can detect danger signals in noisy environments in advance,” adding, “It is also expected to be effective in auditory assistive devices such as artificial cochleae.”
Meanwhile, this research was conducted with the support of the Leader Research program promoted by the Ministry of Science and ICT and the Korea Research Foundation. The research results were published in the international materials science journal Advanced Science (June 17 issue).
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
