A smartphone is all you need for indoor positioning in large spaces.
Prof. Choi Lynn’s team has obtained NET certification for a “deep learning-based geomagnetic indoor positioning technology.”
The technology is planned to be implemented at pilot sites in various indoor spaces such as airports, shopping malls, museums and factories.



▲ Clockwise from the top left: Professor Choi Lynn, Bae Han-june (doctoral student), Ko Young-woong (combined master's and doctoral student) and Son Won-joon (doctoral student).




The deep learning-based geomagnetic indoor positioning technology developed by Professor Choi Lynn’s research team (School of Electrical Engineering, College of Engineering) has obtained New Excellent Technology (NET) certification from the Korean Agency for Technology and Standards under the Ministry of Trade, Industry and Energy. This technology was named a leading new technology at the 2nd New Technology Certification Awards 2020.
* Click to view a video on the technology:

The purpose of NET certification is to accredit the excellence and reliability of new technologies developed in Korea and to promote the commercialization and transfer of those technologies. The certification is conferred by the Korea Industrial Technology Association (KOITA), and the award has been mostly given to conglomerates and SMEs. Professor Choi’s team is the first at the KU Research and Business Foundation to be independently certified for a new technology and is the only university research unit to earn NET certification for two consecutive years (2019 and 2020).

The research team's deep learning-based geomagnetic (terrestrial magnetic) indoor positioning technology is a unique technology developed in Korea. It tracks the indoor location of a person or object by machine learning the distribution pattern of the indoor geomagnetic field using a deep learning recurrent neural network. It is notable that this newly developed indoor positioning system is an innovative world-class technology that greatly improves the performance and economic feasibility by supporting 50 to 80 centimeters of positioning performance in large indoor spaces using only a smartphone, without the need to install additional equipment such as beacons or APs.

To locate positions outdoors, GPS-based locating services are widely used, such as in car navigation systems or smartphone maps. However, a different technology is required for positioning a person or object inside a building because GPS signals cannot be received indoors. Despite research and development over the past 30 years, the most commonly used radio wave-based positioning technology has a large error range (3 to 20 meters), and additional cost is inevitable for the installation and maintenance of beacons or APs. The positioning error range of the system developed by the research team is 73 centimeters, according to an official test conducted by the Korea Laboratory Accreditation Scheme (KOLAS).

The team’s indoor positioning technology was introduced as an example of a participatory smart campus, in which technology developed by school members is applied to a campus, when KU’s SK Future Hall was completed in November 2019. For the first time, location-based services such as indoor map guidance and electronic attendance were applied to an actual campus and could be accessed using only a smartphone without any additional equipment such as beacons. This system is being implemented in two services under construction: location tracking and indoor 3D navigation services for patients and medical staff at Korea University’s Anam Hospital and navigation services for the disabled on KTX, subway lines 1 and 4, the Gyeongui line and at Seoul Station. There are plans to build pilot sites over the next two to three years in various indoor spaces such as airports, shopping malls, museums and factories.

Professor Choi explained: “The indoor positioning technology is a core platform technology of the Fourth Industrial Revolution that enables the development of new location-based services and applications, such as indoor navigation, museum tours, advertisements, logistics, smart safety, augmented reality and customer traffic analysis. It is a state-of-the-art technology that will have a huge ripple effect on the future society and culture comparable to that of semiconductor technology.”



[ Description of Figures ]


 Image 1. Principle of positioning technology based on magnetic vector sequence



Image 2. Process of indoor positioning



Image 3. An example of indoor 3D navigation



Image 4. PCB inside a wearable device which enables indoor positioning without a smartphone



Image 5. Demonstration of indoor positioning at KU’s SK Future Hall


Image 6. Demonstration of 3D indoor navigation at KU’s Hana Square



Image 7. New Excellent Technology (NET) certificate