ICFICE 2019
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
Sequential Location Data Generation Method of RNN Model using Wi-Fi Fingerprint Data |
¿µ¹®Á¦¸ñ(English Title) |
Sequential Location Data Generation Method of RNN Model using Wi-Fi Fingerprint Data |
ÀúÀÚ(Author) |
Hong-Gi Shin
Yong-Hoon Choi
Chang-Pyo Yoon
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 11 NO. 01 PP. 0073 ~ 0075 (2019. 06) |
Çѱ۳»¿ë (Korean Abstract) |
|
¿µ¹®³»¿ë (English Abstract) |
Recently, a method using Wi-Fi fingerprint and deep learning has been studied to provide indoor location-based services more accurately. The RNN model which can memorize past information among the model of deep learning can memorize the continuous movement in the indoor positioning, and can reduce the positioning error. To use the RNN model in Wi-Fifingerprint based indoor positioning, the training data must be sequential. However, the Wi- Fi fingerprint only save the RSSI about the location and cannot be used as training data for the RNN model. In this paper, we propose an area division and movement path generation method that can be used as input data of RNN model based on Wi-Fi fingerprint.
|
Å°¿öµå(Keyword) |
Indoor positioning
Wi-Fi fingerprint
Movement path generation
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|