Ȩ > ÀڷḶ´ç > ÀÚ·á°Ë»ö > Ç¥ÁØ
ÀÚ·á °Ë»ö°á°ú
Ç¥ÁØÁ¾·ù | Á¤º¸Åë½Å±â¼úº¸°í¼(TTAR) | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Ç¥ÁعøÈ£ | TTAR-06.0274 | ±¸ Ç¥ÁعøÈ£ | |||||||||||||||
Á¦°³Á¤ÀÏ | 2023-10-19 | ÃÑ ÆäÀÌÁö | 34 | ||||||||||||||
ÇÑ±Û Ç¥Áظí | ºñ¿¹¾à ±â¹Ý ºñÁ÷±³ ¹«¼± Á¢¼Ó ¹æ½ÄÀ» À§ÇÑ ÀΰøÁö´É Ȱ¿ë (±â¼úº¸°í¼) | ||||||||||||||||
¿µ¹® Ç¥Áظí | Artificial Intelligence (AI) for Grant-free Non-orthogonal Multiple Access (NOMA) (Technical Report) | ||||||||||||||||
ÇÑ±Û ³»¿ë¿ä¾à | ÀÌ ±â¼úº¸°í¼´Â ¹«¼± ¾×¼¼½º¸¦ À§ÇÑ Àΰø Áö´É ¹× ±â°èÇнÀÀÇ »õ·Î¿î Ȱ¿ë »ç·Ê·Î¼, ºñ¿¹¾à ±â¹Ý ºñÁ÷±³ ´ÙÁß Á¢¼Ó ½Ã½ºÅÛÀÇ »óÇ⸵ũ ó¸®À²(throughput)À» ³ôÀ̱â À§ÇÑ ¼Û¼ö½Å ±¸Á¶¸¦ Á¦½ÃÇÑ´Ù. ¿©±â¼, »ç¿ëÀÚÀÇ Àü¼Û °¨Áö¸¦ À§ÇØ ¼ö½ÅµÈ ÇÁ¸®¾Úºí°ú µ¥ÀÌÅ͸¦ ¸ðµÎ Ȱ¿ëÇÏ´Â µ¥ÀÌÅÍ Áö¿ø Ȱ¼º »ç¿ëÀÚ °¨Áö(data-aided active user detection: D-AUD) ¿Í »ç¿ëÀÚ°£ ÄÚµåºÏ Ãæµ¹À» ÀÎÁöÇÏ°í ±¸ºÐÇØ³»´Â Ãæµ¹ ÀÎÁö ´ÙÁß »ç¿ëÀÚ °ËÃâ(collision-aware multi-user detection: CA-MUD) ¹æ½ÄÀÌ °í·ÁµÈ´Ù. D-AUD¿Í CA-MUD¿¡ ÀΰøÁö´É°ú ±â°èÇнÀ ±â¼úÀ» Ȱ¿ëÇÔÀ¸·Î½á Ȱµ¿ °¨Áö ¿À·ùÀ²(activity detection error rate: ADER)¿Í ºñÆ® ¿À·ù·ü(bit-error rate; BER)·Î Á¤ÀÇµÈ Áß°£ ¼º´ÉÁöÇ¥(intermediate KPI)¸¦ °³¼±Çϰí, °á°úÀûÀ¸·Î ÃÖÁ¾ ¼º´ÉÁöÇ¥(eventual KPI)·Î¼ ´ë±Ô¸ð ±â°è Åë½Å(mMTC) ½Ã½ºÅÛ ´ëÇÑ »óÇ⸵ũ ó¸®À² 󸮷®À» Çâ»ó½Ãų ¼ö ÀÖ´Ù. ÀÌ ±â¼úº¸°í¼´Â 3GPP 5G NR ¹«¼±Á¢¼Ó ÀÎÅÍÆäÀ̽º¸¦ À§ÇÑ ÀΰøÁö´É/±â°èÇнÀ¿¡ ´ëÇÑ ±â¼úº¸°í¼ÀÎ TR 38.843ÀÇ ÀϹÝÀûÀÎ ÇÁ·¹ÀÓ¿öÅ©¸¦ ÂüÁ¶ÇÏ¿©, ¹«¼±Á¢¼Ó ÀÎÅÍÆäÀ̽º¿Í ¿¬°èµÈ Ȱ¿ë ¿¹½Ã(use case)¸¦ À§ÇØ ÇÊ¿äÇÑ ¼º´É Æò°¡ ¹æ¹ý·Ð°ú ÀýÂ÷¸¦ ±â¼úÇÑ´Ù. | ||||||||||||||||
¿µ¹® ³»¿ë¿ä¾à | This technical report presents a new application of AI/ML for radio access, proposing a transceiver structure to improve the uplink throughput of grant-free non-orthogonal multiple access(NOMA) systems. In this context, data-aided active user detection (D-AUD) is considered, which utilizes both received preambles and data for user detection, and collision-aware multi-user detection(CA-MUD) is employed to detect codebooks while distinguish their collisions between users. By leveraging AI/ML techniques in D-AUD and CA-MUD, the intermediate key performance indicators (KPIs) defined as activity detection error rate(ADER) and bit-error rate(BER) can be improved, leading to the enhancement of the eventual KPI, specifically the uplink throughput for massive machine-type communication(mMTC) systems. | ||||||||||||||||
°ü·Ã IPR È®¾à¼ | Á¢¼öµÈ IPR È®¾à¼ ¾øÀ½ | ||||||||||||||||
°ü·ÃÆÄÀÏ |
![]() |
||||||||||||||||
Ç¥ÁØÀÌ·Â |
|