| Launch of Safety Sensor Based on Sound Field Change with SecuWorks
| 360-degree detection of unauthorized intrusion and fire situations without blind spots
| Contributing to service application in AI speakers, etc., and localization of sensors

A domestic research team has succeeded in developing a sensor that can detect intrusions or fires using sound.

The Electronics and Telecommunications Research Institute (ETRI) announced on the 3rd that together with Secuworks, it has developed a smart safety sensor that detects intrusions, movement, and even fire by using not only the eyes but also the ears based on changes in the sound field.

The sound field sensor works by generating sound through a speaker and analyzing the changes in the sound field formed in a certain space. When a person moves or the temperature changes, the sound field also changes, and this method detects the situation by receiving the changed sound waves through a microphone.

The newly developed sound field sensor consists of three major parts: a microphone, a speaker, and a signal processing unit. It is 8 cm x 5 cm in size and can be attached to ceilings and other places to capture subtle sounds and movements.

For example, if you call an AI speaker and set it to security mode, the speaker will emit a sound similar to the chirping of crickets for 0.5 seconds every 2 to 3 seconds. It actively and periodically sends sound waves to identify the sound field created in the space. If a change in the sound field is detected due to movement or temperature, a text message or other notification is sent to the user.

The biggest advantage of the new sensor is that there are no blind spots. Existing image sensors or infrared sensors could not detect or gave false alarms in places that were not visible, such as heat shields. However, the sound field sensor uses the phenomenon of sound reflection and diffraction to sensitively detect movement in blind spots, even over obstacles.

In addition, the initial fire situation can be quickly identified. Unlike existing sensors that detect temperature changes only after the fire has spread significantly, the sound field sensor can detect changes within 50 seconds even in blind spots.

The sound field sensor developed by the research team was released as a modular product in the form of a speaker and microphone integration (HW) and a sound field signal processing chip. It can also be serviced through software updates for existing CCTVs and AI speakers. It is easy to install and has high expandability with IoT devices, etc.

Since there are no blind spots, the same space can be detected with a smaller number of sensors than existing sensors. Therefore, it is advantageous in that it can be installed at about 30% of the cost of existing sensors. The installation and maintenance costs are also lower than other sensors, and the accuracy is high, reducing unnecessary false calls.

The developed sensor is expected to be widely used in places where theft prevention and fire safety are required, such as the rapidly increasing number of single-person households and public facilities. Furthermore, it seems that it can be applied to welfare care services, such as detecting the movements of the elderly and providing notifications.

This product was developed by transferring the original sound field security sensor technology from Dr. Kang-Ho Park's team at ETRI's Intelligent Sensor Research Lab. From 2010 to the present, the research team has applied for and registered 10 patents related to sound field sensor technology in five countries: Korea, the United States, Germany, the United Kingdom, and China.

There have been studies in the past on intrusion detection using standing wave sounds.

However, since it mainly detected a single frequency in an analog manner, it was difficult to accurately distinguish changes in the sound field.

On the other hand, the research team processed sounds of various frequencies into digital signals in the form of a sound spectrum and analyzed them with an intelligent algorithm to accurately distinguish and identify movement and temperature changes.

Dr. Park Kang-ho, head of research at ETRI, said, “Currently, Japan accounts for more than half of the global sensor market, including thermal imaging cameras,” and “If this sensor is commercialized, it will supplement existing sensors, replace imports, and help domestically produce sensor products and material components.”

SecureworkVice President Kim Seung-joo also said, “In addition to the standalone product released this time, we plan to release a modular product in the third quarter of this year and sell it to leading domestic and foreign companies. We are also conducting joint research and development with KT.”

The full-scale product launch is scheduled for September of this year. Sales through trial production are expected to begin next year. Secuworks is also exploring global expansion, including to Silicon Valley in the United States.

In the future, the research team plans to develop sensors that utilize inaudible sound waves, which are sounds that humans cannot hear, and conduct research to more accurately distinguish between types of movement and temperature changes through deep learning. They are also planning to expand their business areas to areas such as smart home appliances, smart cities, port containers, and ships.

Meanwhile, according to Technavio, the global market for intelligent video security and black boxes is expected to grow to $18.8 billion by 2024.