Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
ÇѱÛÁ¦¸ñ(Korean Title) |
IOT ȯ°æ¿¡¼ÀÇ ¿ÀÅäÀÎÄÚ´õ ±â¹Ý Ư¡ ÃßÃâÀ» ÀÌ¿ëÇÑ ³×Æ®¿öÅ© ħÀÔŽÁö ½Ã½ºÅÛ |
¿µ¹®Á¦¸ñ(English Title) |
Network Intrusion Detection System Using Feature Extraction Based on AutoEncoder in IOT environment |
ÀúÀÚ(Author) |
Joohwa Lee
Keehyun Park
ÀÌÁÖÈ
¹Ú±âÇö
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 08 NO. 12 PP. 0483 ~ 0490 (2019. 12) |
Çѱ۳»¿ë (Korean Abstract) |
³×Æ®¿öÅ© ħÀÔ Å½Áö ½Ã½ºÅÛ(NIDS)¿¡¼ ºÐ·ùÀÇ ±â´ÉÀº »ó´çÈ÷ Áß¿äÇϸç ŽÁö ¼º´ÉÀº ´Ù¾çÇÑ Æ¯Â¡¿¡ µû¶ó ´Þ¶óÁø´Ù. ÃÖ±Ù µö·¯´×¿¡ ´ëÇÑ ¿¬±¸°¡ ¸¹ÀÌ ÀÌ·ç¾îÁö°í ÀÖÀ¸³ª ³×Æ®¿öÅ© ħÀÔŽÁö ½Ã½ºÅÛ¿¡¼´Â ¸¹Àº ¼öÀÇ Æ®·¡ÇÈ°ú °íÂ÷¿øÀÇ Æ¯Â¡À¸·Î ÀÎÇÏ¿© ¼Óµµ°¡ ´À·ÁÁö´Â ¹®Á¦Á¡ÀÌ ÀÖ´Ù. µû¶ó¼ µö·¯´×À» ºÐ·ù¿¡ »ç¿ëÇÏ´Â °ÍÀÌ ¾Æ´Ï¶ó Ư¡ ÃßÃâÀ» À§ÇÑ Àüó¸® °úÁ¤À¸·Î »ç¿ëÇϸç ÃßÃâÇÑ Æ¯Â¡À» ±â¹ÝÀ¸·Î ºÐ·ùÇÏ´Â ¿¬±¸ ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. µö·¯´×ÀÇ ´ëÇ¥ÀûÀÎ ºñÁöµµ ÇнÀÀÎ Stacked AutoEncoder¸¦ »ç¿ëÇÏ¿© Ư¡À» ÃßÃâÇÏ°í Random Forest ºÐ·ù ¾Ë°í¸®ÁòÀ» »ç¿ëÇÏ¿© ºÐ·ùÇÑ °á°ú ºÐ·ù ¼º´É°ú ŽÁö ¼ÓµµÀÇ Çâ»óÀ» È®ÀÎÇÏ¿´´Ù. IOT ȯ°æ¿¡¼ ¼öÁýÇÑ µ¥ÀÌÅ͸¦ ÀÌ¿ëÇÏ¿© Á¤»ó ¹× °ø°ÝÆ®·¡ÇÈÀ» ¸ÖƼŬ·¡½º·Î ºÐ·ùÇÏ¿´À» ¶§ 99% ÀÌ»óÀÇ ¼º´ÉÀ» º¸¿´À¸¸ç, AE-RF, Single-RF¿Í °°Àº ´Ù¸¥ ¸ðµ¨°ú ºñ±³ÇÏ¿´À» ¶§µµ ¼º´É ¹× ŽÁö¼Óµµ°¡ ¿ì¼öÇÑ °ÍÀ¸·Î ³ªÅ¸³µ´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
In the Network Intrusion Detection System (NIDS), the function of classification is very important, and detection performance depends on various features. Recently, a lot of research has been carried out on deep learning, but network intrusion detection system experience slowing down problems due to the large volume of traffic and a high dimensional features. Therefore, we do not use deep learning as a classification, but as a preprocessing process for feature extraction and propose a research method from which classifications can be made based on extracted features. A stacked AutoEncoder, which is a representative unsupervised learning of deep learning, is used to extract features and classifications using the Random Forest classification algorithm. Using the data collected in the IOT environment, the performance was more than 99% when normal and attack traffic are classified into multiclass, and the performance and detection rate were superior even when compared with other models such as AE-RF and Single-RF.
|
Å°¿öµå(Keyword) |
NIDS
IOT
Unsupervised Learning
Machine Learning
AutoEncoder
³×Æ®¿öÅ© ħÀÔŽÁö½Ã½ºÅÛ
»ç¹°ÀÎÅͳÝ
ºñÁöµµÇнÀ
±â°èÇнÀ
¿ÀÅäÀÎÄÚ´õ
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|