• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Current Result Document : 34 / 36

ÇѱÛÁ¦¸ñ(Korean Title) ´ÙÃø¸é ±â°èÇнÀÀ» »ç¿ëÇÑ ½º¸¶Æ® À̵¿ °´Ã¼ÀÇ À§Ä¡ º¸Á¤ ±â¹ý
¿µ¹®Á¦¸ñ(English Title) A Location Compensation Method for Smart Mobile Objects using Multilateral Machine Learning
ÀúÀÚ(Author) ±Ç¿ëÈÆ   Á¤Àιü   Yonghun Kwon   Inbum Jung  
¿ø¹®¼ö·Ïó(Citation) VOL 26 NO. 07 PP. 0312 ~ 0321 (2020. 07)
Çѱ۳»¿ë
(Korean Abstract)
»ç¹°ÀÎÅÍ³Ý È¯°æ¿¡¼­ À§Ä¡ ±â¹Ý ¼­ºñ½º¸¦ À§ÇÑ ±â¼ú Áß¿¡¼­ ºí·çÅõ½º ºñÄÜÀº È°¹ßÇÏ°Ô ¿¬±¸µÇ°í ÀÖ´Â ±â¼úÀÌ´Ù. ºñÄÜ¿¡¼­ ¹ß»êµÇ´Â ½ÅÈ£ Áß RSSI °ªÀ» ÀÌ¿ëÇÏ¿© ½Ç³»¿¡¼­ À̵¿ ÁßÀÎ ½º¸¶Æ® °´Ã¼¿Í »ç¹° °£ÀÇ °Å¸®ÃøÁ¤À» À¯µµÇÒ ¼ö ÀÖ´Ù. ±×·¯³ª RSSI´Â ½ÅÈ£ÀÇ ¹Ý»ç¿Í ȸÀý°ú °°Àº Çö»ó¿¡ ÀÇÇØ ¿µÇâÀ» ¹Þ¾Æ Á¤È®µµ°¡ ³ôÁö ¾ÊÀº ¹®Á¦Á¡À» °¡Áö°í ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ´ÙÃø¸é ±â°èÇнÀ ¾Ë°í¸®ÁòµéÀ» »ç¿ëÇÏ¿© RSSI¸¦ »ç¿ëÇÑ °Å¸®ÃøÁ¤ÀÇ Á¤È®µµ¸¦ ³ôÀÌ´Â ¿¬±¸¸¦ ÁøÇàÇÏ¿´´Ù. ÀúÁÖÆÄ Åë°ú ÇÊÅÍ¿Í ÆĶó¹ÌÅÍ ÇнÀ ±×¸®°í ¿ªÀüÆÄ ÇнÀÀ» ¼øÂ÷ÀûÀ¸·Î ÁøÇàÇÏ¿© °Å¸®¸¦ ÃøÁ¤ÇÏ¿´´Ù. Á¦¾ÈµÈ ´ÙÃø¸é ±â°èÇнÀ ¹æ½ÄÀÇ ¼º´É Æò°¡¸¦ À§ÇØ ½ÇÇèÀÇ °¢ ´Ü°è¿¡¼­ À¯µµµÈ °Å¸®¿Í ½ÇÁ¦ °Å¸® °£ÀÇ ¿ÀÂ÷¸¦ ¹ÙÅÁÀ¸·Î Á¤È®µµ¸¦ Æò°¡ÇÏ¿´´Ù. ½ÇÇè °á°ú Á¦¾ÈµÈ °Å¸®ÃøÁ¤ ¹æ½ÄÀº Á¦ÇÑµÈ °Å¸® ¹üÀ§ ³»¿¡¼­ ½ÇÃø°ª¿¡ ±ÙÁ¢ÇÑ °Å¸® °ªÀ» À¯µµÇÔÀ» º¸À̹ǷÎ, »ç¹°ÀÎÅÍ³Ý ¸Á¿¡¼­ À§Ä¡ ÀÎ½Ä ¼­ºñ½º ÀÀ¿ëÇÁ·Î±×·¥¿¡ ±â¿©ÇÒ ¼ö ÀÖÀ½À» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Bluetooth technology has been widely used for location awareness services in IoT network environments. RSSI emitted from a beacon can be utilized to measure the distance between smart mobile objects that are moving indoors. However, reflection and diffraction phenomena that affect RSSI often result in inaccurate distance measurements. In this paper, multilateral machine learning algorithms are applied to enhance the accuracy of distance measurements based on RSSI. First, a low-pass filter is used to correct estimates that overshoot among the measured RSSI data. Second, a parameter control leaning method is used and the distance translation equation is supplied with corrected RSSI data. Although this equation can predict distance from the strength of the RSSI signal it can generate errors when it comes to the range of the distances estimated. To reduce these range of distance errors, a back propagation neural network algorithm is applied. For performance evaluation, we compare the actual distance values with the estimated distance values that come from each stage during the multilateral machine learning. The estimated values were found to be close to the actual distances within the limited scope of distances tested, as such, we believe this technology can contribute to location awareness service applications in IoT networks.
Å°¿öµå(Keyword) ´ÙÃø¸é ÇнÀ   ÀúÁÖÆÄ Åë°ú ÇÊÅÍ   ÆĶó¹ÌÅÍ ÇнÀ   ½Å°æ¸Á   ½º¸¶Æ® ¸ð¹ÙÀÏ °´Ã¼   multilateral learning   low pass filter   parameter learning   neural network   smart mobile object  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå