• ๋Œ€ํ•œ์ „๊ธฐํ•™ํšŒ
Mobile QR Code QR CODE : The Transactions of the Korean Institute of Electrical Engineers
  • COPE
  • kcse
  • ํ•œ๊ตญ๊ณผํ•™๊ธฐ์ˆ ๋‹จ์ฒด์ด์—ฐํ•ฉํšŒ
  • ํ•œ๊ตญํ•™์ˆ ์ง€์ธ์šฉ์ƒ‰์ธ
  • Scopus
  • crossref
  • orcid

  1. (Department of Electrical Engineering, Hanbat National University, Korea)



Reflectometry, Propagation constant, Instantaneous frequency estimation Algorithm, Simulator, Time-Frequency domain Reflectometry

1. ์„œ ๋ก 

์ „๋ ฅ ์‹œ์Šคํ…œ์˜ ๊ด€๋ฆฌ๋ฅผ ์œ„ํ•ด ์ผ€์ด๋ธ” ๊ณ ์žฅ์  ํƒ์ง€ ๋ฐ ๋…ธํ™” ์ธก์ •์— ๋Œ€ํ•œ ๋ฐฉ๋ฒ• ๊ฐœ๋ฐœ์ด ํ•„์š”ํ•˜๋‹ค. ์ตœ๊ทผ ์žฅ๊ฑฐ๋ฆฌ ์†ก์ „์˜ ํ•„์š”์„ฑ์ด ์ฆ๊ฐ€ํ•˜๋ฉด์„œ HVDC(High Voltage Direct Current)์˜ ์‹œ์žฅ๊ทœ๋ชจ๊ฐ€ ์„ฑ์žฅํ•˜๊ณ  ์žˆ๋‹ค. HVDC ์ผ€์ด๋ธ”์€ ์šฉ๋Ÿ‰์„ฑ ์ถฉ์ „์ „๋ฅ˜๊ฐ€ ์—†์–ด ์žฅ๊ฑฐ๋ฆฌ ์†ก์ „์— ์œ ๋ฆฌํ•˜๊ณ  ์œ ์ „์ฒด, ์‹œ์Šค ์†์‹ค์ด ์—†์–ด ๋Œ€์šฉ๋Ÿ‰ ์†ก์ „์ด ๊ฐ€๋Šฅํ•˜๋‹ค. ๋˜ํ•œ ๋„์ฒด ์ €ํ•ญ์ด ๋‚ฎ์•„ ์†ก์ „ ์†์‹ค์ด ์ ๊ณ , ์ „๋ ฅ์กฐ๋ฅ˜ ๋ฐ ๋ถ€ํ•˜ ์ œ์–ด๊ฐ€ ์ˆ˜์›”ํ•˜์—ฌ ์„ ๋กœ ์ด์šฉ๋ฅ ์„ ๋†’์ผ ์ˆ˜ ์žˆ๋Š” ์žฅ์ ์ด ์žˆ๋‹ค (1). ์ด์— ๋”ฐ๋ผ ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์— ์ ์šฉ ๊ฐ€๋Šฅํ•œ ์ง„๋‹จ ๋ฐฉ๋ฒ•์˜ ๊ฐœ๋ฐœ์ด ์š”๊ตฌ๋˜๊ณ  ์žˆ๋‹ค.

ํ˜„์žฌ ์ผ€์ด๋ธ”์˜ ๊ณ ์žฅ์  ํƒ์ง€ ๋ฐ ๊ฑด์ „์„ฑ ํ‰๊ฐ€์— ์ „๊ธฐ์ ์ธ ๋ฐฉ๋ฒ•์œผ๋กœ๋Š” VLF(Very Low frequency) tan delta ์ธก์ •๋ฒ•, PD(Partial Discharge)์ง„๋‹จ๋ฒ•์ด ๋Œ€ํ‘œ์ ์œผ๋กœ ์ ์šฉ๋˜๊ณ  ์žˆ๋‹ค (2). VLF tan delta๋Š” ๊ทน์ €์ฃผํŒŒ ์ „์•• ์†Œ์Šค๋ฅผ ์ธ๊ฐ€ํ•œ ํ›„์— tan delta๋ฅผ ์ธก์ •ํ•˜๋Š” ์ง„๋‹จ๋ฒ•์œผ๋กœ, tan delta๋Š” ์ ˆ์—ฐ์ฒด์˜ ์†์‹ค์— ์˜ํ•œ ์†์‹ค์ „๋ฅ˜์™€ ์ปคํŒจ์‹œํ„ฐ์˜ ์ „๋ฅ˜์˜ ๋น„๋ฅผ ๋‚˜ํƒ€๋‚ด๊ณ , tan delta๊ฐ€ ํด์ˆ˜๋ก ๋ˆ„์„ค์ „๋ฅ˜๊ฐ€ ์ฆ๊ฐ€ํ•˜๋Š” ๊ฒƒ์„ ์˜๋ฏธํ•œ๋‹ค. tan delta๋Š” ๋…ธ์ด์ฆˆ์˜ ์˜ํ–ฅ์— ๊ฐ•์ธํ•ด ์ •๋ฐ€ํ•œ ํ˜„์žฅ ์ง„๋‹จ์ด ๊ฐ€๋Šฅํ•˜๊ณ  ์ผ€์ด๋ธ”์˜ ์ „์ฒด์ ์ธ ์—ดํ™” ์ƒํƒœ๋ฅผ ์ง„๋‹จํ•  ์ˆ˜ ์žˆ์œผ๋ฉฐ IEEE std, 400.2 ์ง€์นจ์„œ์— ํŒ์ • ๊ธฐ์ค€์ด ์ œ์‹œ๋˜์–ด ์žˆ๋‹ค (3). ๊ทธ๋Ÿฌ๋‚˜ ์ผ€์ด๋ธ” ๊ธธ์ด์— ๋”ฐ๋ผ ์ฆ๊ฐ€ํ•˜๋Š” ์ปคํŒจ์‹œํ„ฐ ์ถฉ์ „์šฉ๋Ÿ‰์œผ๋กœ ์ธํ•ด 1.2 km ์ด์ƒ์˜ ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์—๋Š” ์ ์šฉ์ด ์–ด๋ ต๋‹ค (4). PD๋ž€ ๊ตญ๋ถ€์  ๋ฐฉ์ „ํ˜„์ƒ์œผ๋กœ ์ ˆ์—ฐ์ฒด ํ‘œ๋ฉด์ด๋‚˜ ๋‚ด๋ถ€์˜ ๊ณต๊ทน์— ์˜ํ•ด ๋ฐœ์ƒํ•œ๋‹ค. ์ด๋Š” ์ ˆ์—ฐ์ฒด์˜ ์—ดํ™”๋ฅผ ์ผ์œผํ‚ค๋Š” ํŒŒ๊ดด์˜ ์ฃผ์›์ธ์ด๋‹ค. PD ์ง„๋‹จ๋ฒ•์€ ๋ถ€๋ถ„ ๋ฐฉ์ „ ์‹œ ๋ฐœ์ƒํ•˜๋Š” ํŽ„์Šค๋ฅผ ํ†ตํ•ด ๊ณ ์žฅ์ ์„ ์ธก์ •ํ•  ์ˆ˜ ์žˆ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ๋ถ€๋ถ„๋ฐฉ์ „์ด ์ผ์–ด๋‚œ ์ง€์ ์—์„œ์˜ ๋ฐœ์ƒํ•œ ํŽ„์Šค๋Š” ์ผ€์ด๋ธ”์„ ๋”ฐ๋ผ ์ด๋™ํ•˜๋ฉฐ ๊ฐ์‡ ์™€ ๋ถ„์‚ฐ์„ ๋น ๋ฅด๊ฒŒ ๊ฒช๊ฒŒ ๋œ๋‹ค. ๋”ฐ๋ผ์„œ ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์˜ ๊ณ ์žฅ์  ํƒ์ง€๋ฅผ ์œ„ํ•ด์„œ๋Š” ์ถ”๊ฐ€์ ์ธ ์‹ ํ˜ธ ์ฒ˜๋ฆฌ ๊ธฐ์ˆ  ๊ฐœ๋ฐœ์ด ํ•„์š”ํ•˜๋‹ค.

๊ณ ์žฅ์  ํƒ์ง€์— ์ƒˆ๋กญ๊ฒŒ ๋„์ž…๋˜๊ณ  ์žˆ๋Š” ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์€ ์ž„ํ”ผ๋˜์Šค ๋ถˆ์ผ์น˜ ์ง€์ ์—์„œ ๋ฐ˜์‚ฌ๋œ ์‹ ํ˜ธ๋ฅผ ๋ถ„์„ํ•˜์—ฌ ๊ณ ์žฅ์ ์„ ํƒ์ง€ํ•˜๋Š” ๊ธฐ์ˆ ์ด๋‹ค. ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์€ PLC ํ†ต์‹ ์šฉ์œผ๋กœ ๊ฐœ๋ฐœ๋œ ๋น„์ ‘์ด‰์‹ ์ปคํ”Œ๋Ÿฌ๋ฅผ ์ด์šฉํ•˜์—ฌ ํ™œ์„  ์ƒํƒœ์˜ ์ง„๋‹จ์ด ๊ฐ€๋Šฅํ•˜๋ฉฐ ์ธ๊ฐ€ํ•œ ์‹ ํ˜ธ์— ๋”ฐ๋ผ ํฌ๊ฒŒ ์‹œ๊ฐ„ ์˜์—ญ, ์ฃผํŒŒ์ˆ˜ ์˜์—ญ, ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์˜ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์œผ๋กœ ๋ถ„๋ฅ˜ํ•  ์ˆ˜ ์žˆ๋‹ค (5).

TDR(Time Domain Reflectometry)์€ ์‹œ๊ฐ„ ์˜์—ญ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์œผ๋กœ ๊ณ„๋‹จ ์‹ ํ˜ธ ๋˜๋Š” ํŽ„์Šค ์‹ ํ˜ธ๋ฅผ ๊ธฐ์ค€ ์‹ ํ˜ธ๋กœ ์‚ฌ์šฉํ•œ๋‹ค (6). ์ž„ํ”ผ๋˜์Šค ๋ถˆ์ผ์น˜ ์ง€์ ์—์„œ ๊ธ‰๊ฒฉํ•œ ์ „์••์˜ ๋ณ€ํ™”๊ฐ€ ์ƒ๊ธด ๋ฐ˜์‚ฌ ์‹ ํ˜ธ ๋ฐœ์ƒ์ง€์ ์„ ๊ณ ์žฅ์ ์œผ๋กœ ํŒ๋‹จํ•œ๋‹ค.

FDR(Frequency Domain Reflectometry)์€ ์ฃผํŒŒ์ˆ˜ ์˜์—ญ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์œผ๋กœ ์œ„์ƒ์„ ๊ฒ€์ถœํ•˜๋Š” PD-FDR์„ ์ฃผ๋กœ ์‚ฌ์šฉํ•œ๋‹ค (7). ์ •ํ˜„ํŒŒ๋ฅผ ๊ธฐ์ค€ ์‹ ํ˜ธ๋กœ ์‚ฌ์šฉํ•˜๋ฉฐ ์ฃผํŒŒ์ˆ˜๋ฅผ ์ฆ๊ฐ€์‹œํ‚ค๋ฉฐ ์ผ€์ด๋ธ”์— ์ธ๊ฐ€ํ•œ๋‹ค. ์ด๋•Œ ์ €์ฃผํŒŒ ํ†ต๊ณผ ํ•„ํ„ฐ๋ฅผ ํ†ตํ•ด DC ์„ฑ๋ถ„๋งŒ ์ถ”์ถœํ•œ ์‹ ํ˜ธ์˜ ์œ„์ƒ์„ ๊ฒ€์ถœํ•œ๋‹ค. ์ž„ํ”ผ๋˜์Šค ๋ถˆ์ผ์น˜ ์ง€์ ์—์„œ ์œ„์ƒ ์ฐจ์ด๋ฅผ ๊ณ„์‚ฐํ•˜์—ฌ ๊ณ ์žฅ์  ๊ฑฐ๋ฆฌ๋ฅผ ์ธก์ •ํ•œ๋‹ค (7).

TFDR(Time-Frequency Domain Reflectometry)๋Š” ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์˜ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์œผ๋กœ ์‹œ๊ฐ„์— ๋”ฐ๋ผ ์ฃผํŒŒ์ˆ˜๊ฐ€ ์„ ํ˜•์ ์œผ๋กœ ์ฆ๊ฐ€ํ•˜๋Š” ์ฒฉ ์‹ ํ˜ธ๋ฅผ ๊ธฐ์ค€ ์‹ ํ˜ธ๋กœ ์‚ฌ์šฉํ•˜๊ณ  ์‹œ๊ฐ„ ํญ(Time Duration, TD), ์ฃผํŒŒ์ˆ˜ ๋Œ€์—ญํญ(Frequency Band width, BW), ์ค‘์‹ฌ ์ฃผํŒŒ์ˆ˜(Center Frequency, CF)๋ฅผ ์กฐ์ ˆํ•˜์—ฌ ์‹ ํ˜ธ๋ฅผ ์„ค๊ณ„ํ•  ์ˆ˜ ์žˆ๋‹ค. ์ธ๊ฐ€ ์‹ ํ˜ธ์™€ ์ทจ๋“ํ•œ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ๋ฅผ ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์ƒํ˜ธ์ƒ๊ด€ํ•จ์ˆ˜(Time-Frequency Cross Correlation, TFCC)๋ฅผ ํ†ตํ•ด ๊ณ ์žฅ์ ์„ ํƒ์ง€ํ•  ์ˆ˜ ์žˆ์œผ๋ฉฐ, ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์—์„œ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋ฅผ ํ†ตํ•ด ์‹ ํ˜ธ ํ•ด์„์ด ๊ฐ€๋Šฅํ•˜๋‹ค.

์ „๊ธฐ์  ์‹ ํ˜ธ ํ•ด์„์„ ๊ธฐ๋ฐ˜์œผ๋กœ ํ•œ ์ง„๋‹จ๋ฒ•์€ ๊ณตํ†ต์œผ๋กœ ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ ์ง„๋‹จ์„ ์œ„ํ•ด ์‹ ํ˜ธ์˜ ๊ฐ์‡ ๋ฅผ ํ•ด๊ฒฐํ•ด์•ผ ํ•œ๋‹ค. TFDR์€ ์ธ๊ฐ€ ์‹ ํ˜ธ์™€ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ์˜ ์œ ์‚ฌ๋„๋ฅผ ์ด์šฉํ•œ TFCC๋ฅผ ํ†ตํ•ด ๊ณ ์žฅ์ ์„ ํ™•์ธํ•œ๋‹ค. ๋”ฐ๋ผ์„œ ์‹ ํ˜ธ์˜ ํฌ๊ธฐ๋ฅผ ์ด์šฉํ•˜๋Š” ๋‹ค๋ฅธ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•๊ณผ๋Š” ๋‹ค๋ฅด๊ฒŒ ์‹ ํ˜ธ๊ฐ€ ๊ฐ์‡ ๋ฅผ ๊ฒช๋”๋ผ๋„ ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์—์„œ ๊ณ ์žฅ์  ํƒ์ง€๊ฐ€ ๊ฐ€๋Šฅํ•˜๋‹ค. ์ด๋Š” ์‹ค์ œ ์ œ์ฃผ-ํ•ด๋‚จ HVDC ์„ ๋กœ์— ์ ์šฉ๋˜๊ณ  ์žˆ์–ด ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์˜ ๊ณ ์žฅ์  ํƒ์ง€ ๊ธฐ์ˆ ๋กœ ์ฃผ๋ชฉ๋ฐ›๊ณ  ์žˆ๋‹ค (1). ๊ทธ๋Ÿฌ๋‚˜ ์„ค๊ณ„๋˜๋Š” ๊ธฐ์ค€ ์‹ ํ˜ธ๋Š” ๋‹ค์–‘ํ•œ ์ฃผํŒŒ์ˆ˜๊ฐ€ ํฌํ•จ๋œ ์‹ ํ˜ธ์ด๊ธฐ ๋•Œ๋ฌธ์— ์ „ํŒŒ ์‹œ ์ฃผํŒŒ์ˆ˜์— ๋”ฐ๋ฅธ ์†๋„์˜ ์ฐจ์ด๊ฐ€ ๋ฐœ์ƒํ•œ๋‹ค. ๋˜ํ•œ ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ฆ๊ฐ€ํ• ์ˆ˜๋ก ์‹ ํ˜ธ๋Š” ์ฃผํŒŒ์ˆ˜์— ์˜์กดํ•œ ๊ฐ์‡ ๋กœ ์‹ ํ˜ธ๊ฐ€ ํฌํ•จํ•˜๊ณ  ์žˆ๋Š” ์ฃผํŒŒ์ˆ˜ ์„ฑ๋ถ„๋“ค์˜ ์†๋„๊ฐ€ ๋‹ฌ๋ผ์ง€๋ฉฐ, ๊ฐ์‡ ๊ฐ€ ์ผ์–ด๋‚˜๋Š” ๋ถ„์‚ฐ์ด ๋ฐœ์ƒํ•œ๋‹ค. ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์ฃผํŒŒ์ˆ˜ ๋ณ„ ์„ฑ๋ถ„์ด ์ „ํŒŒ ๊ฑฐ๋ฆฌ์— ๋”ฐ๋ผ ์™œ๊ณก์ด ๋ฐœ์ƒํ•˜๊ฒŒ ๋˜๋ฉด ๊ฒฐํ•จ์—์„œ ์ƒ์„ฑ๋œ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ์™€ ์ธ๊ฐ€ ์‹ ํ˜ธ์˜ TFCC ๊ฐ’์ด ๊ฐ์†Œ๋˜์–ด ์ธ๊ฐ€ ์‹ ํ˜ธ์™€ ์œ ์‚ฌ์„ฑ์ด ๋–จ์–ด์ง€๊ณ  ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์—์„œ ๊ณ ์žฅ์  ํƒ์ง€์˜ ์„ฑ๋Šฅ์ด ์ €ํ•˜๋œ๋‹ค.

๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก ์‹œ ์ „ํŒŒ ๊ฑฐ๋ฆฌ์— ๋”ฐ๋ผ ์ฆ๊ฐ€ํ•˜๋Š” ๊ฐ์‡  ๋ฐ ๋ถ„์‚ฐ์„ ๊ณ ๋ คํ•˜์—ฌ, ์ฃผํŒŒ์ˆ˜์™€ ์ „ํŒŒ ๊ฑฐ๋ฆฌ์— ๋”ฐ๋ฅธ ์ „ํŒŒ ์†๋„์˜ ๊ฒฝํ–ฅ์„ฑ์„ ๋ถ„์„ํ•˜๊ณ ์ž ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋ฅผ ํ†ตํ•ด ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์—์„œ ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜๋ฅผ ๋ถ„์„ ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ๊ตฌํ˜„ํ•˜๊ณ  ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์— ์ ์šฉํ•˜๊ธฐ ์œ„ํ•ด ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋ฅผ ๊ฐœ๋ฐœํ•˜๊ณ  ์ œ์•ˆํ•œ ์•Œ๊ณ ๋ฆฌ์ฆ˜ ์„ฑ๋Šฅ์„ ๊ฒ€์ฆํ•˜์˜€๋‹ค.

2. ๋ณธ ๋ก 

2.1 ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•

๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์ด๋ž€ ์ž„ํ”ผ๋˜์Šค ๋ถˆ์ผ์น˜ ์ง€์ ์—์„œ ๋Œ์•„์˜ค๋Š” ๋ฐ˜์‚ฌํŒŒ๋ฅผ ์ธก์ •ํ•˜์—ฌ ๊ณ ์žฅ์ ์„ ํƒ์ง€ํ•˜๋Š” ๊ธฐ๋ฒ•์ด๋‹ค. ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์—์„œ ์‚ฌ์šฉํ•˜๋Š” ๊ธฐ์ค€ ์‹ ํ˜ธ๋Š” ๊ฐ€์šฐ์‹œ์•ˆ ํฌ๋ฝ์„  ์„ ํ˜• ์ฒฉ ์‹ ํ˜ธ์ด๋ฉฐ, ์‹œ๊ฐ„์— ๋”ฐ๋ผ ์ฃผํŒŒ์ˆ˜๊ฐ€ ์„ ํ˜•์ ์œผ๋กœ ๋ณ€ํ™”ํ•œ๋‹ค. ์‹ ํ˜ธ๋Š” ๋‹ค์Œ ์‹ (1)๊ณผ ๊ฐ™๋‹ค (8).

(1)
$s(t)=e^{-\dfrac{A(t-t_{0})^{2}}{2}+\dfrac{j B(t-t_{0})^{2}}{2}+j\omega_{0}(t-t_{0})}$

($A=\dfrac{1}{2\tau_{0}^{2}}$, $B=\sqrt{(2\pi BW)^{2}2A-A^{2}}$, $w_{0}=2\pi f_{0}$ )

$\tau_{0}$๋Š” ์‹œ๊ฐ„ ํญ(Time Duration)์˜ ์˜ํ–ฅ์„ ๋ฐ›์œผ๋ฉฐ ์‹ ํ˜ธ ๋Œ€ ์žก์Œ๋น„์— ๋น„๋ก€ํ•œ๋‹ค. $f_{0}$, $t_{0}$๋Š” ์ค‘์‹ฌ ์ฃผํŒŒ์ˆ˜(Center Frequency), ์ค‘์‹ฌ ์‹œ๊ฐ„์„ ์˜๋ฏธํ•œ๋‹ค. BW๋Š” ์ฃผํŒŒ์ˆ˜ ๋Œ€์—ญํญ(Bandwidth)๋ฅผ ์˜๋ฏธํ•˜๋ฉฐ, ์‹ ํ˜ธ์˜ ๋ถ„ํ•ด๋Šฅ(Resolution)์€ ๋‚˜์ดํ€ด์ŠคํŠธ ์ด๋ก (Nyquist theorem)์— ๋”ฐ๋ผ ์ฃผํŒŒ์ˆ˜ ๋Œ€์—ญํญ์— ๋น„๋ก€ํ•˜๋ฉฐ ๋‹ค์Œ ์‹ (2)์™€ ๊ฐ™๋‹ค (9).

(2)
$\Delta R=\dfrac{v_{p}}{2BW}$

๊ฐ€์šฐ์‹œ์•ˆ ํฌ๋ฝ์„  ์„ ํ˜• ์ฒฉ ์‹ ํ˜ธ๋Š” ์‹œ๊ฐ„ ํญ๊ณผ ์ฃผํŒŒ์ˆ˜ ๋Œ€์—ญํญ์ด ๋™์‹œ์— ์ฆ๊ฐ€ํ•  ์ˆ˜ ์—†๋Š” ๋ถˆํ™•์ •์„ฑ ์›๋ฆฌ๋ฅผ ๊ณ ๋ คํ•˜์—ฌ ์ ์ ˆํžˆ ์‹ ํ˜ธ๋ฅผ ์„ค๊ณ„ํ•ด์•ผ ํ•˜๋ฉฐ ๋ถˆํ™•์ •์„ฑ ์›๋ฆฌ ํŒ๋ณ„์‹์€ ๋‹ค์Œ๊ณผ ๊ฐ™๋‹ค (10).

(3)
$TB=\dfrac{2\piยท\dfrac{BW}{\sqrt{2}}}{6}ยท\dfrac{\dfrac{TD}{\sqrt{2}}}{6}\ge 0.5$

์‹ (2), (3)์— ๋”ฐ๋ผ ๊ธฐ์ค€ ์‹ ํ˜ธ ์„ค๊ณ„ ์‹œ ๋ถ„ํ•ด๋Šฅ๊ณผ ๋ถˆํ™•์ •์„ฑ์˜ ์›๋ฆฌ๋ฅผ ๋™์‹œ์— ๊ณ ๋ คํ•˜์—ฌ ์‹ ํ˜ธ๋ฅผ ์„ค๊ณ„ํ•ด์•ผ ํ•œ๋‹ค. ์„ค๊ณ„๋œ ์‹ ํ˜ธ๋Š” ๋‹ค์–‘ํ•œ ์ฃผํŒŒ์ˆ˜ ์„ฑ๋ถ„์œผ๋กœ ๊ตฌ์„ฑ๋˜๊ณ , ์ฃผํŒŒ์ˆ˜์— ๋”ฐ๋ฅธ ์ „ํŒŒ ์†๋„๊ฐ€ ์ฐจ์ด๊ฐ€ ์žˆ์–ด ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ๋ฉ€์–ด์งˆ์ˆ˜๋ก ์‹ ํ˜ธ๋Š” ๋ถ„์‚ฐ ๋ฐ ์™œ๊ณก๋œ๋‹ค. ๋”ฐ๋ผ์„œ ์ „ํŒŒ ์ƒ์ˆ˜ ๋ฐ ๊ฑฐ๋ฆฌ์— ๋”ฐ๋ผ ์˜ํ–ฅ์„ ๋ฐ›๋Š” ์‹ ํ˜ธ ๋ถ„์„์ด ํ•„์š”ํ•˜๋‹ค.

์ „ํŒŒ๋œ ์‹ ํ˜ธ๋Š” ์ „ํŒŒ ๋งค์งˆ์˜ ์ „ํŒŒ ์ƒ์ˆ˜๋ฅผ ๊ณ ๋ คํ•œ ์ฃผํŒŒ์ˆ˜ ์ข…์† ํ•„ํ„ฐ $G(z,\:w)= e^{-jk(w)z}$์— ์˜ํ•ด ์‹ ํ˜ธ๊ฐ€ ์™œ๊ณก๋˜๋ฉฐ, ์ด๋Š” ๋‹ค์Œ ์‹ (4)์™€ ๊ฐ™๋‹ค. ์ฃผํŒŒ์ˆ˜ ์˜์—ญ์˜ ์‹ (4)๋ฅผ ํ‘ธ๋ฆฌ์— ์—ญ๋ณ€ํ™˜ํ•˜์—ฌ ์‹œ๊ฐ„ ์˜์—ญ์—์„œ ์ „ํŒŒ ๊ฑฐ๋ฆฌ $z$ ๋งŒํผ ์ „ํŒŒ๋œ ์‹ ํ˜ธ๋ฅผ ํ™•์ธํ•  ์ˆ˜ ์žˆ๊ณ  ์ด๋Š” ๋‹ค์Œ ์‹ (5)์™€ ๊ฐ™๋‹ค (11).

(4)
$S(z,\:w)=e^{-jk(w)z}S(0,\:w)$

(5)
$s(z,\:t)=\dfrac{1}{2\pi}\int_{-\infty}^{\infty}e^{j(wt-k(w)z)}S(0,\:w)dw$

์ „ํŒŒ ์ƒ์ˆ˜ $k$๋Š” $\beta -j\alpha$๋กœ ๊ตฌ์„ฑ๋˜๋ฉฐ ์‹ค์ˆ˜๋ถ€ $\beta$๋Š” ์œ„์ƒ ์ƒ์ˆ˜, ํ—ˆ์ˆ˜๋ถ€ $\alpha$๋Š” ๊ฐ์‡  ์ƒ์ˆ˜์ด๋‹ค. ์‹ ํ˜ธ๊ฐ€ ์ข์€ ์ฃผํŒŒ์ˆ˜ ๋Œ€์—ญ์„ ๊ฐ–๋Š” ํ˜‘๋Œ€์—ญ ์‹ ํ˜ธ(narrow band signal)์ผ ๋•Œ, $\omega_{0}$์— ๋Œ€ํ•ด ํ…Œ์ผ๋Ÿฌ ๊ธ‰์ˆ˜๋กœ ์ด๊ณ„๋„ํ•จ์ˆ˜๊นŒ์ง€ ํ™•์žฅํ•˜์—ฌ ์ถ”์ •ํ•˜๋ฉด ๋‹ค์Œ ์‹ (6)๊ณผ ๊ฐ™๋‹ค (12).

(6)
$\left .\left . k_{0}=k(\omega_{0}),\:\dot k_{0}=\dfrac{dk}{d\omega}\right |_{w_{0}},\:\ddot k_{0}=\dfrac{d^{2}k}{d\omega^{2}}\right |_{w_{0}}$

(7)
$k(\omega)=k_{0}+\dot k_{0}(\omega -\omega_{0})+\dfrac{1}{2}\ddot k_{0}(\omega -\omega_{0})^{2}+\cdots$

์—ฌ๊ธฐ์„œ $\dot k_{0}=\dot\beta_{0}-j\dot\alpha_{0}$์ด๋ฉฐ $\dot\beta_{0}$์€ ์ „ํŒŒ ์†๋„์˜ ์—ญ์ˆ˜๋ฅผ ๋‚˜ํƒ€๋‚ธ๋‹ค. $k(\omega)$๋ฅผ ๊ณ ๊ณ„๋กœ ์ถ”์ •ํ• ์ˆ˜๋ก ์ •๋ฐ€ํ•œ ํ‘œํ˜„์ด ๊ฐ€๋Šฅํ•˜์ง€๋งŒ ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ๊ฐ์‡  ๋ฐ ๋ถ„์‚ฐ์˜ ์›์ธ์ด ๋˜๋Š” ์ด๊ณ„๋„ํ•จ์ˆ˜ $\ddot k_{0}=\ddot\beta_{0}-j\ddot\alpha_{0}$๊นŒ์ง€ ๋‚˜ํƒ€๋ƒˆ๋‹ค. ์ „ํŒŒ ์ƒ์ˆ˜๋ฅผ ์ด๊ณ„๋„ํ•จ์ˆ˜๊นŒ์ง€ ๊ณ ๋ คํ•˜์—ฌ ๊ธฐ์ค€ ์‹ ํ˜ธ๋ฅผ ๋‚˜ํƒ€๋‚ด๋ฉด ๋‹ค์Œ ์‹ (8)๊ณผ ๊ฐ™๋‹ค.

(8)
$s(z,\:t)=\dfrac{1}{2\pi}\int_{-\infty}^{\infty}e^{jw(t-k_{0}'z)-jk_{0}''zw/2}S(0,\:w)dw$

์ „ํŒŒ ์ƒ์ˆ˜์˜ ์ด๊ณ„๋„ํ•จ์ˆ˜์— ๋”ฐ๋ผ ์‹ ํ˜ธ์˜ ์‹œ๊ฐ„์— ๋”ฐ๋ฅธ ์ฃผํŒŒ์ˆ˜ ์ฆ๋ถ„ $\dot\omega_{0}$๋Š” ๋‹ค์Œ ์‹ (9)์™€ ๊ฐ™๋‹ค.

(9)
$\dot\omega_{0}=\dfrac{\ddot\beta_{0}z}{(\tau_{0}^{2}+\ddot\alpha_{0})^{2}+(\ddot\beta_{0}z)^{2}}$

์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ฆ๊ฐ€ํ•จ์— ๋”ฐ๋ผ TD์˜ ์˜ํ–ฅ์„ ๋ฐ›๋Š” $\tau_{0}$์—๋„ ๋ณ€์กฐ๊ฐ€ ์ ์šฉ๋˜๋ฉฐ, ์ด๋Š” $\tau_{c hi rp}$์œผ๋กœ ํ‘œํ˜„ํ•  ์ˆ˜ ์žˆ๋‹ค. $\tau_{ch i rp}^{2}$์€ $\tau_{0}$์™€ ์ฃผํŒŒ์ˆ˜ ์ฆ๋ถ„ $\dot\omega_{0}$์˜ ๋ณต์†Œ์ˆ˜ ํ˜•ํƒœ๋กœ ๋‹ค์Œ ์‹ (10)๊ณผ ๊ฐ™๋‹ค.

(10)
$\tau_{ch i rp}^{2}=\dfrac{\tau_{0}^{2}(1+j\dot\omega_{0}\tau_{0}^{2})}{1+j\dot\omega_{0}\tau_{0}^{4}}$

์œ„ ์‹์— ๋”ฐ๋ผ ์ฒ˜ํ•‘์ด ์ ์šฉ๋œ ์ „ํŒŒ ์‹ ํ˜ธ์™€ ์ŠคํŽ™ํŠธ๋Ÿผ์€ ์‹์€ ๋‹ค์Œ๊ณผ ๊ฐ™๋‹ค.

(11)
$S(z,\:\omega)=\sqrt{2\pi\tau_{ch i rp}^{2}}e^{-j\dot k_{0}z\omega}e^{-(\tau_{ch i rp}^{2}+ j\ddot k_{0}z)\omega^{2}/2}$

(12)
$S(z,\:t)=\sqrt{\dfrac{\tau_{ch i rp}^{2}}{\tau_{ch i rp}^{2}+ j\ddot k_{0}z}}\exp\left[-(t-\dot k_{0}z)^{2}\dfrac{}{2(\tau_{ch i rp}^{2}+j\ddot k_{0}z)}\right]$

์‹ ํ˜ธ์˜ ๋ถ„์‚ฐ์„ ๊ณ ๋ คํ•  ๊ฒฝ์šฐ ๋ถ„์‚ฐ์„ ๋‚˜ํƒ€๋‚ด๋Š” ์ด๊ณ„ ์ „ํŒŒ ์ƒ์ˆ˜ ๊ฐ’์„ ๊ณ ๋ คํ•œ ๊ทธ๋ฃน ์‹œ๊ฐ„ ์ง€์—ฐ ์‹์„ ์ด์šฉํ•˜์—ฌ ๊ทธ๋ฃน ์†๋„๋ฅผ ๊ตฌํ•˜๋ฉฐ ๋‹ค์Œ ์‹ (13)๊ณผ ๊ฐ™๋‹ค.

(13)
$t_{g}=\dot\beta_{0}z-\dfrac{\dot\alpha_{0}\ddot\beta_{0}z^{2}}{\tau_{0}^{2}+\ddot\alpha_{0}z}$

๊ทธ๋Ÿฌ๋‚˜ ์ด๋Š” ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ถฉ๋ถ„ํžˆ ์ฆ๊ฐ€ํ•  ๊ฒฝ์šฐ ๊ณ ์ฃผํŒŒ ๋ถ€๋ถ„์ด ์†Œ๋ฉธ๋˜๋Š” ๊ฒƒ์„ ๊ณ ๋ คํ•˜์ง€ ์•Š์•˜๊ธฐ ๋•Œ๋ฌธ์— ์ „ํŒŒ ์‹ ํ˜ธ์˜ ๋ถ„์‚ฐ์„ ๋ณด์ •ํ•˜๋Š” ํ•„ํ„ฐ ์„ค๊ณ„ ๊ฐœ๋ฐœ์ด ํ•„์š”ํ•˜๋‹ค.

2.2 ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜ ์ถ”์ • ์•Œ๊ณ ๋ฆฌ์ฆ˜

์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜(Instantaneous Frequency, IF)๋ž€ ์ฃผํŒŒ์ˆ˜๊ฐ€ ์‹œ๊ฐ„์— ๋”ฐ๋ผ ๋ณ€ํ™”ํ•  ๋•Œ ํŠน์ • ์‹œ๊ฐ„์—์„œ์˜ ์ฃผํŒŒ์ˆ˜๋กœ ์ •์˜๋˜๋ฉฐ ์‹ ํ˜ธ์˜ ์‹œ๊ฐ„์— ๋Œ€ํ•œ ๋ณ€ํ™” ํŠน์„ฑ์˜ ์„ค๋ช…์„ ์œ„ํ•ด ๋งŽ์ด ์‚ฌ์šฉ๋œ๋‹ค (13). ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜๋Š” ํž๋ฒ„ํŠธ ๋ณ€ํ™˜์„ ์ด์šฉํ•˜์—ฌ ์‹ ํ˜ธ์˜ ์‹ค์ˆ˜๋ถ€ $s(t)$์™€ ํ—ˆ์ˆ˜๋ถ€ $\hat s(t)$๋ฅผ ๊ฐ–๋Š” ํ•ด์„์  ์‹ ํ˜ธ๋กœ ๋งŒ๋“  ๋’ค IF๋ฅผ ๋‹ค์Œ ์‹ (15)๋กœ ํ‘œํ˜„ํ•  ์ˆ˜ ์žˆ๋‹ค.

(14)
$\varphi(t)=\arg(s(t)+j\hat s(t))$

(15)
$f_{s}=\dfrac{1}{2\pi}\dfrac{d\varphi}{dt}$

์ „ํŒŒ๋œ ์‹ ํ˜ธ๋Š” ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ(Wigner-Ville Distribution)๋กœ ๋ณ€ํ™˜ํ•˜์—ฌ ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์˜ ํŠน์ง•์„ ๋™์‹œ์— ๊ด€์ฐฐํ•  ์ˆ˜ ์žˆ๋‹ค. ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋Š” ์‹ ํ˜ธ์˜ ์—๋„ˆ์ง€ ๋ถ„ํฌ๋ฅผ ๋‚˜ํƒ€๋‚ด๋ฉฐ ๋‹ค์Œ ์‹ (16)๊ณผ ๊ฐ™๋‹ค (14).

(16)
$W_{s}(t,\:\omega)=\dfrac{1}{2\pi}\int_{-\infty}^{\infty}s(t+\dfrac{1}{2}\tau)s^{*}(t-\dfrac{1}{2}\tau)e^{-j\tau\omega}d\tau$

์ „ํŒŒ ์‹ ํ˜ธ์—์„œ ๋ฐœ์ƒํ•œ ์™œ๊ณก์€ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์—์„œ ๋‚˜ํƒ€๋‚œ ๊ฐ ์‹ ํ˜ธ์˜ IF ์ถ”์ •์„ ํ†ตํ•ด ํ™•์ธํ•  ์ˆ˜ ์žˆ์ง€๋งŒ, ์žฅ๊ฑฐ๋ฆฌ ์ „ํŒŒ๋œ ์‹ ํ˜ธ๋Š” ๊ฐ์‡ ๊ฐ€ ์‹ฌํ•ด ์‹ ํ˜ธ์˜ ์กด์žฌ๋ฅผ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์—์„œ ํŒŒ์•…ํ•˜๊ธฐ ์–ด๋ ต๋‹ค (15).

IF ์ถ”์ •์€ ์–ด๋ ต์ง€๋งŒ TFCC๋ฅผ ํ†ตํ•ด ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ์œ ์‚ฌํ•œ ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์œ„์น˜๋ฅผ ํƒ์ง€ํ•  ์ˆ˜ ์žˆ์œผ๋ฉฐ ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ์˜ ํฌ๊ธฐ๋กœ ๋‚˜๋ˆ ์ฃผ๊ธฐ ๋•Œ๋ฌธ์— ์‹ ํ˜ธ์˜ ํฌ๊ธฐ์— ์ƒ๊ด€์—†์ด ๊ฒฐ๊ณผ๋ฅผ ๋„์ถœํ•  ์ˆ˜ ์žˆ๋‹ค. ์ •๊ทœํ™”๋œ ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ์ „ํŒŒ ์‹ ํ˜ธ์˜ TFCC๋Š” ์‹ (17)๊ณผ ๊ฐ™๋‹ค (16).

(17)
$C_{sr}(t)=\dfrac{1}{E_{s}E_{r}(t)}\iint W_{r}(t',\:w)W_{s}(t'-t,\:w)dwdt'$

(18)
$E_{r}(t)=\iint W_{r}(t',\:w)dwdt'$

(19)
$E_{s}(t)=\iint W_{s}(t ,\:w)dwdt$

์—ฌ๊ธฐ์„œ, $E_{s}(t)$, $E_{r}(t)$๋Š” ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ์˜ ์—๋„ˆ์ง€๋ฅผ ๋‚˜ํƒ€๋‚ด๋ฉฐ, TFCC๋Š” ๋‘ ์‹ ํ˜ธ์˜ ์—๋„ˆ์ง€๋ฅผ ๋‚˜๋ˆ„์–ด ์ •๊ทœํ™”ํ•˜๊ธฐ ๋•Œ๋ฌธ์— ๋ฐ˜์‚ฌ ์‹ ํ˜ธ์˜ ํฌ๊ธฐ์™€ ๊ด€๊ณ„์—†์ด 0๊ณผ 1 ์‚ฌ์ด์˜ ์œ ์‚ฌ๋„๋ฅผ ๊ฐ–๊ฒŒ ๋œ๋‹ค.

ํ˜‘๋Œ€์—ญ ์‹ ํ˜ธ์— ๊ฐ€๊นŒ์šธ์ˆ˜๋ก ํž๋ฒ„ํŠธ ๋ณ€ํ™˜๋œ ์‹ ํ˜ธ๋Š” ์ง๊ต ์‹ ํ˜ธ์— ๊ฐ€๊นŒ์›Œ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์—์„œ ๊ฐ€์žฅ ๋†’์€ ์—๋„ˆ์ง€์—์„œ IF๋ฅผ ํšจ๊ณผ์ ์œผ๋กœ ์ถ”์ •ํ•  ์ˆ˜ ์žˆ๋‹ค. ๋˜ํ•œ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์˜ ํ”ผํฌ ๊ธฐ๋ฐ˜ IF ์ถ”์ •์€ ์„ ํ˜• FM ์‹ ํ˜ธ์— ์ตœ์ ํ™”๋œ ๋ฐฉ๋ฒ•์ด๋‹ค (17). ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ•์—์„œ๋Š” ์‹œ๊ฐ„์— ๋”ฐ๋ผ ์ฃผํŒŒ์ˆ˜๊ฐ€ ์„ ํ˜•์ ์œผ๋กœ ์ฆ๊ฐ€ํ•˜๋Š” ์‹ ํ˜ธ๋ฅผ ์‚ฌ์šฉํ•˜๋ฏ€๋กœ ์ฃผํŒŒ์ˆ˜ ์˜์—ญ์—์„œ IF๋Š” ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์—์„œ ๊ฐ€์žฅ ๋†’์€ ์—๋„ˆ์ง€๋ฅผ ๊ฐ–๋Š” ์ ์„ ์˜๋ฏธํ•œ๋‹ค.

๊ทธ๋ฆผ. 1. ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ ๋ฉ”์ธ ํ™”๋ฉด

Fig. 1. Simulator main screen

../../Resources/kiee/KIEE.2023.72.6.757/fig1.png

ํ•˜์ง€๋งŒ ์‹ ํ˜ธ๋Š” ๊ฐ์‡ ๊ฐ€ ์‹ฌํ•ด ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋ฅผ ํ†ตํ•œ IF ์ธก์ •์ด ์–ด๋ ค์›Œ ์‹ ํ˜ธ ๋ณต์› ์•Œ๊ณ ๋ฆฌ์ฆ˜์ด ํ•„์š”ํ•˜๋‹ค. ์ด์— ๋”ฐ๋ผ min-max ์ •๊ทœํ™”๋ฅผ ์ด์šฉํ•˜์—ฌ ์ „ํŒŒ ์‹ ํ˜ธ์˜ ํฌ๊ธฐ๋ฅผ ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ๋™์ผํ•˜๊ฒŒ 0๊ณผ 1 ์‚ฌ์ด ๊ฐ’์œผ๋กœ ๋ณต์›ํ•˜์˜€์œผ๋ฉฐ ๋‹ค์Œ ์‹ (20)์™€ ๊ฐ™๋‹ค.

(20)
$x_{no}{al}=\dfrac{{x}-{x}_{\min}}{{x}_{\max}-{x}_{\min}}$

๋ณต์›๋œ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์˜ ํ”ผํฌ์ ์„ ํ†ตํ•ด IF ์ถ”์ • ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ๊ตฌํ˜„ํ•˜์˜€๋‹ค, IF๋ฅผ 1์ฐจ ํ•จ์ˆ˜๋กœ ์ถ”์ •ํ•˜๊ณ  ์„ ํ˜• ์ตœ์†Œ์ œ๊ณฑ๋ฒ•(Linear Least Squares Estimation, LLS)์œผ๋กœ ํ‘œํ˜„ํ•˜์˜€์œผ๋ฉฐ ๋‹ค์Œ ์‹(21)๊ณผ ๊ฐ™๋‹ค.

(21)
$J=\sum_{i=1}^{m}(y_{i}-x_{i}^{T}w)^{2}=(Y-Xw)^{T}(Y-Xw)$

(22)
$Y=Xw$, $Y=\begin{bmatrix}y_{1}\\y_{2}\\โ€ค\\โ€ค\\โ€ค\\y_{n}\end{bmatrix}$, $X=\begin{bmatrix}x_{1}&1\\x_{2}&1\\โ€ค&.\\โ€ค&.\\โ€ค&.\\x_{n}&1\end{bmatrix}$, $w=\begin{bmatrix}a\\b\end{bmatrix}$

์—ฌ๊ธฐ์„œ Y๋Š” ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜๋ฅผ 1์ฐจ ํ•จ์ˆ˜๋กœ ํ‘œํ˜„ํ•˜๊ธฐ ์œ„ํ•ด ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์˜ ์ œ์ผ ํฐ ๊ฐ’๋“ค์„ ๋‚จ๊ฒจ๋†“์€ ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ํ‰๋ฉด์˜ ์ฃผํŒŒ์ˆ˜๋ฅผ ๋‚˜ํƒ€๋‚ด๋ฉฐ, X๋Š” ์‹œ๊ฐ„ ์ถ•์˜ ๋ฐ์ดํ„ฐ์˜ ์ž…๋ ฅ ๋ฒกํ„ฐ์™€ ๊ฐ’๋“ค๊ณผ ํฌ๊ธฐ๊ฐ€ n์ด๊ณ  ๋ชจ๋“  ์š”์†Œ๊ฐ€ 1์ธ ์—ด๋ฒกํ„ฐ๋ฅผ ๋‚˜๋ž€ํžˆ ๋†“์€ ๊ฒƒ์ด๋‹ค. $w$๋Š” ๊ฐ€์ค‘์น˜ ๋ฒกํ„ฐ๋ฅผ ์˜๋ฏธํ•˜๋ฉฐ ์—ฌ๊ธฐ์„œ $w$๋Š” ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜์˜ 1์ฐจ ํ•จ์ˆ˜ํ‘œํ˜„ $y=ax+b$์˜ ๊ธฐ์šธ๊ธฐ์™€ $y$์ ˆํŽธ์„ ๋‚˜ํƒ€๋‚ธ๋‹ค. J๋Š” ๋น„์šฉํ•จ์ˆ˜๋ฅผ ๋‚˜ํƒ€๋‚ด๋ฉฐ J๋ฅผ $w$์— ๋Œ€ํ•ด ๋ฏธ๋ถ„ํ•˜์—ฌ ๊ฐ€์ค‘์น˜ $w$ ๊ตฌํ•  ์ˆ˜ ์žˆ์œผ๋ฉฐ ๋‹ค์Œ ์‹ (24)์™€ ๊ฐ™๋‹ค.

(23)
$\dfrac{\partial J}{\partial w}=2X^{T}Xw-X^{T}Y=0$

(24)
$w=(X^{T}X)^{-1}X^{T}Y$

๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์— LLS์„ ํ†ตํ•ด ์ถ”์ •ํ•œ 1์ฐจ ํ•จ์ˆ˜์˜ ๊ธฐ์šธ๊ธฐ๋ฅผ ๋ฐ˜์˜ํ•˜์˜€๋‹ค. TFCC ๊ธฐ๋ฐ˜์˜ ์‹ ํ˜ธ ๋ณต์› ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ํ™œ์šฉํ•˜์—ฌ ์‹ ํ˜ธ๋ฅผ ๋ณต์›ํ•˜๊ณ  LLS๋ฅผ ํ†ตํ•ด IF๋ฅผ ์ถ”์ •ํ•  ์ˆ˜ ์žˆ๋Š” ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ๊ตฌํ˜„ํ•˜์˜€๋‹ค.

3. ์‹คํ—˜ ๊ฒฐ๊ณผ ๋ฐ ๊ณ ์ฐฐ

3.1 ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ ๊ฐœ๋ฐœ

HVDC ์‹œ์žฅ ๊ทœ๋ชจ์˜ ํ™•๋Œ€์— ๋”ฐ๋ผ ์žฅ๊ฑฐ๋ฆฌ ์†ก์ „์˜ ํ•„์š”์„ฑ์ด ์ฆ๊ฐ€ํ•˜๊ณ  ์žˆ๋‹ค. ํ•˜์ง€๋งŒ HVDC ์ผ€์ด๋ธ” ๊ณ ์žฅ์  ํƒ์ง€๋Š” ์ „๋ฌธ๊ฐ€์˜ ์ˆ™๋ จ๋„ ๋ฐ ๊ฒฝํ—˜์— ์˜์กด๋˜๋ฉฐ, ์ •ํ™•ํ•œ ์ง„๋‹จ์ด ๋ถˆ๊ฐ€๋Šฅํ•œ ์‚ฌ๋ก€๊ฐ€ ๋งŽ๋‹ค (1). ์ด์— ๋”ฐ๋ผ ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์— ์ ์šฉ ๊ฐ€๋Šฅํ•œ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ• ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋ฅผ ๊ฐœ๋ฐœํ•˜์˜€๋‹ค. ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์˜ ๊ฒฝ์šฐ ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ฆ๊ฐ€ํ• ์ˆ˜๋ก ์‹ ํ˜ธ์˜ ๊ฐ์‡ ์™€ ๋ถ„์‚ฐ์ด ์‹ฌํ•ด์ ธ ๊ณ ์žฅ์  ํƒ์ง€ ์„ฑ๋Šฅ ์ €ํ•˜๋˜๋ฉฐ, IF ์ถ”์ •์„ ํ†ตํ•œ ์ฃผํŒŒ์ˆ˜ ์„ฑ๋ถ„์˜ ๊ฐ์‡  ํŠน์„ฑ ๊ธฐ๋ฐ˜ ๋ณด์ • ํ•„ํ„ฐ ์„ค๊ณ„๊ฐ€ ํ•„์š”ํ•˜๋‹ค. ์ด๋ฅผ ์œ„ํ•ด์„œ IF ์ถ”์ •์ด ํ•„์š”ํ•˜์ง€๋งŒ, ์žฅ๊ฑฐ๋ฆฌ ์ „ํŒŒ ์‹ ํ˜ธ๋Š” ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์—์„œ ํƒ์ง€๊ฐ€ ์–ด๋ ต๋‹ค. ์‹ ํ˜ธ์˜ ์กด์žฌ ์—ฌ๋ถ€๋Š” TFCC๊ฐ€ ํ•ด๊ฒฐํ•ด์ค„ ์ˆ˜ ์žˆ์ง€๋งŒ, ์‹ ํ˜ธ ์—๋„ˆ์ง€ ์ž์ฒด๊ฐ€ ์ž‘์•„์ ธ ์žˆ๋Š” ์ƒํ™ฉ์—์„œ IF ์ถ”์ •์€ ๋ถˆ๊ฐ€๋Šฅํ•˜๋‹ค. ๋”ฐ๋ผ์„œ ๋ณธ ๋…ผ๋ฌธ์—์„œ ์ „ํŒŒ ๊ฑฐ๋ฆฌ์— ๋”ฐ๋ฅธ ์™œ๊ณก์„ ๋ฐ˜์˜ํ•œ ์ „ํŒŒ ์‹ ํ˜ธ์™€ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ๋ฅผ ๋ถ„์„ ๊ฐ€๋Šฅํ•œ ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋ฅผ ๊ฐœ๋ฐœํ•˜์˜€๊ณ , ์ œ์•ˆํ•œ ์‹ ํ˜ธ ๋ณต์› ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ํ†ตํ•ด ์ „ํŒŒ ์‹ ํ˜ธ๋ฅผ ๊ธฐ์ค€ ์‹ ํ˜ธ์˜ ํฌ๊ธฐ์— ๋งž๊ฒŒ ๋ณต์›ํ•˜๊ณ  IF ์ถ”์ • ์•Œ๊ณ ๋ฆฌ์ฆ˜์˜ ์„ฑ๋Šฅ์„ ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋ฅผ ํ†ตํ•ด ๊ฒ€์ฆํ•˜์˜€๋‹ค.

์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋Š” ์‹ ํ˜ธ ์ธ๊ฐ€๋ถ€, ์‹ ํ˜ธ ๊ณ„์ธก๋ถ€, ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌํ™”๋ฉด ๋ฐ ์ „ํŒŒ ์ƒ์ˆ˜ ์ž…๋ ฅ๋ถ€, IF ๊ณ„์ธก๋ถ€๋กœ ๊ตฌ์„ฑ๋œ๋‹ค. ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ์˜ ๋ฉ”์ธ ํ™”๋ฉด์€ ๊ทธ๋ฆผ 1๊ณผ ๊ฐ™๋‹ค.

(1) ์‹ ํ˜ธ ์ธ๊ฐ€๋ถ€

โž€ ์˜์—ญ ์„ค์ •(Domain) : ๊ณ„์ธก๋ฒ• ์„ ํƒ์˜์—ญ์œผ๋กœ ์‹œ๊ฐ„ ์˜์—ญ, ์ฃผํŒŒ์ˆ˜ ์˜์—ญ, ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์„ ๊ฒฐ์ •ํ•œ๋‹ค.

โž ๊ธฐ๋ณธ๊ฐ’(Default) : ์ž„์˜๋กœ ์„ค์ •ํ•œ ์‹ ํ˜ธ ์ƒ์„ฑ์— ํ•„์š”ํ•œ ๊ฐ’์„ ๋ถˆ๋Ÿฌ์˜จ๋‹ค. ์‹ค์ œ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก ์‹œ ์‚ฌ์šฉ๋˜๋Š” ์žฅ๋น„๋ฅผ ๊ณ ๋ คํ•˜์—ฌ ์„ค์ •ํ•˜์˜€๋‹ค.

โž‚ ๋ถˆํ™•์ •์„ฑ ์›๋ฆฌ ํ™•์ธ(Check) : ๋ถˆํ™•์ •์„ฑ์˜ ์›๋ฆฌ์— ๋”ฐ๋ผ ์‹œ๊ฐ„ ํญ๊ณผ ์ฃผํŒŒ์ˆ˜ ๋Œ€์—ญํญ์˜ ๊ณฑ์€ ๋ฐ˜๋“œ์‹œ 0.5์ด์ƒ์˜ ๊ฐ’์„ ๋งŒ์กฑํ•ด์•ผ ํ•œ๋‹ค. ์ด์— ๋”ฐ๋ผ ์ค‘์‹ฌ ์ฃผํŒŒ์ˆ˜๋ฅผ ๊ธฐ์ค€์œผ๋กœ ์ƒ˜ํ”Œ๋ง ๋ ˆ์ดํŠธ(Sampling Rate)๊ฐ€, ์ฃผํŒŒ์ˆ˜ ๋Œ€์—ญํญ์„ ๊ธฐ์ค€์œผ๋กœ ์‹œ๊ฐ„ ํญ์ด ๋ณ€๊ฒฝ๋œ๋‹ค.

โžƒ ์‹ ํ˜ธ ์ƒ์„ฑ(Signal Generation) : ์„ค์ •๋œ TD, BW, CF ๊ฐ’์„ ๋ฐ˜์˜ํ•˜์—ฌ ๊ธฐ์ค€ ์‹ ํ˜ธ๋ฅผ ์ƒ์„ฑํ•œ๋‹ค.

(2) ์‹ ํ˜ธ ๊ณ„์ธก๋ถ€

โž„ ๊ธฐ๋ณธ๊ฐ’(Default) : ์ „ํŒŒ ์‹ ํ˜ธ์˜ ๊ณ„์ธก์ง€์ ๊ณผ ๋ฐ์ดํ„ฐ ์ทจ๋“์„ ์œ„ํ•œ ์„ค์ •๊ฐ’์„ ๋ถˆ๋Ÿฌ์˜จ๋‹ค.

โž… ๋ฐ์ดํ„ฐ ๊ณ„์ธก ๋ฐ ์ทจ๋“(Data Acquisition) : ์‹ ํ˜ธ๋ฅผ ๊ณ„์ธกํ•œ ํ›„ ๋ฐ์ดํ„ฐ๋ฅผ ์ทจ๋“ํ•˜์—ฌ ์ธ๊ฐ€ ์‹ ํ˜ธ ๋ฐ ์ „ํŒŒ ์‹ ํ˜ธ๋ฅผ ๋‚˜ํƒ€๋‚ด๊ณ , ์ด์— ๋”ฐ๋ฅธ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ ๊ฐ’์„ ์ทจ๋“ํ•œ๋‹ค.

โž† ์ „ํŒŒ ์ƒ์ˆ˜ : ๋งค์งˆ์˜ ํŠน์„ฑ์ธ ์ „ํŒŒ ์ƒ์ˆ˜๋ฅผ ์ž…๋ ฅํ•  ์ˆ˜ ์žˆ์œผ๋ฉฐ, ์‹ค์ˆ˜ ๋ถ€๋ถ„๊ณผ ํ—ˆ์ˆ˜ ๋ถ€๋ถ„์œผ๋กœ ๋‚˜๋ˆ„์–ด์ ธ ์žˆ๋‹ค. ์ „ํŒŒ ์†๋„๋Š” $\dot\beta_{0}$์˜ ์—ญ์ˆ˜๋กœ ํ‘œํ˜„ํ•œ๋‹ค.

โž‡ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ : ์ธ๊ฐ€ ์‹ ํ˜ธ์™€ ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋ฅผ ๋‚˜ํƒ€๋‚ธ๋‹ค.

โžˆ ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜ : ์ธ๊ฐ€ ์‹ ํ˜ธ์™€ ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜ ๊ธฐ์šธ๊ธฐ๋ฅผ on/off๋กœ ํ‘œ์‹œํ•  ์ˆ˜ ์žˆ๋‹ค. ์‹ ํ˜ธ๊ฐ€ ์ „ํŒŒ ๋ ์ˆ˜๋ก ๊ฐ์‡  ๋ฐ ๋ถ„์‚ฐ์„ ๊ฒช์–ด ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์—์„œ IF ์ถ”์ •์ด ์–ด๋ ค์›Œ ์‹ ํ˜ธ ๋ณต์› ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ํ†ตํ•ด ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ๊ฐ™์€ ์—๋„ˆ์ง€ ๋ถ„ํฌ๋กœ ์กฐ์ •ํ•˜๊ณ  ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜๋ฅผ ํ‘œ์‹œํ•˜์˜€๋‹ค.

3.2 ์‹คํ—˜ ๊ตฌ์„ฑ ๋ฐ ๊ฒฐ๊ณผ

๋ณธ ๋…ผ๋ฌธ์—์„œ ์ œ์•ˆํ•œ ์•Œ๊ณ ๋ฆฌ์ฆ˜์˜ ์„ฑ๋Šฅ์„ ๊ฒ€์ฆํ•˜๊ธฐ ์œ„ํ•ด ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ• ์‹œ์Šคํ…œ ์„ค๊ณ„๋ฅผ ๊ณ ๋ คํ•˜์—ฌ ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋ฅผ ๊ฐœ๋ฐœํ•˜์˜€๊ณ  ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ• ์‹œ์Šคํ…œ ๊ตฌ์„ฑ๋„๋Š” ๊ทธ๋ฆผ 2์™€ ๊ฐ™๋‹ค. ์‹ ํ˜ธ๋Š” ๋น„์ ‘์ด‰์‹ ์ปคํ”Œ๋Ÿฌ๋ฅผ ํ†ตํ•ด ์ธ๊ฐ€ํ•˜๊ณ  ๊ณ„์ธกํ•˜๋ฉฐ, ๋™์ผ ์œ„์น˜์—์„œ ์‹ ํ˜ธ ์ธ๊ฐ€ ๋ฐ ์ทจ๋“์„ ๊ฐ€์ •ํ•˜์—ฌ ๊ฐœ๋ฐœํ•˜์˜€๋‹ค. ์‹ ํ˜ธ๋ฐœ์ƒ๊ธฐ์—์„œ ์ƒ์„ฑ๋œ ๊ธฐ์ค€ ์‹ ํ˜ธ๋Š” T-์ปค๋„ฅํ„ฐ๋ฅผ ํ†ตํ•ด ๋””์ง€ํ„ธ ์˜ค์‹ค๋กœ์Šค์ฝ”ํ”„์— ์ทจ๋“๋˜๊ณ  ์‹ ํ˜ธ์ฒ˜๋ฆฌ ์‹œ์Šคํ…œ์— ์ „์†ก๋œ๋‹ค. ๋‹ค๋ฅธ ์ฑ„๋„์„ ํ†ตํ•ด ์„ค๊ณ„๋œ ๊ธฐ์ค€ ์‹ ํ˜ธ๋Š” ์ผ€์ด๋ธ”์— ์ธ๊ฐ€๋˜๊ณ  ์ž„ํ”ผ๋˜์Šค ๋ถˆ์ผ์น˜ ์ง€์ ์—์„œ ๋ฐœ์ƒํ•œ ์‹ ํ˜ธ๋Š” ๋ฐ˜์‚ฌ๋˜์–ด ๋Œ์•„์˜จ๋‹ค. ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ์ „ํŒŒ์— ๋Œ€ํ•œ ์‹ ํ˜ธ ๊ฐ์‡  ๋ฐ ๋ถ„์‚ฐ์„ ๋ถ„์„ํ•˜๊ธฐ ์œ„ํ•ด์„œ ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋ฅผ ๊ฐœ๋ฐœํ•˜์˜€๊ณ , ์‹œ๋ฎฌ๋ ˆ์ด์…˜์€ ์ „ํŒŒ ์‹ ํ˜ธ ์ธก์ •๊ณผ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ ์ธก์ • ๋‘ ์ƒํ™ฉ์œผ๋กœ ์‹ ํ˜ธ ๋ถ„์„์ด ๊ฐ€๋Šฅํ•˜๋‹ค. ๋ฐ˜์‚ฌ ์‹ ํ˜ธ ์ธก์ • ์‹œ ์ผ์ • ์ง€์ ์— ๋กœ๋“œ ์ž„ํ”ผ๋˜์Šค๋ฅผ ๋ฌดํ•œ๋Œ€๋กœ ์ž…๋ ฅํ•˜์—ฌ ์ข…๋‹จ์„ ๋ชจ์˜ํ•  ์ˆ˜ ์žˆ๋‹ค.

์ทจ๋“ํ•œ ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ๋ฐ˜์‚ฌ ์‹ ํ˜ธ์˜ TFCC๋ฅผ ํ†ตํ•ด ์ž„ํ”ผ๋˜์Šค ๋ถˆ์ผ์น˜ ์ง€์ ์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๊ณ  ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋ฅผ ํ†ตํ•ด ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์—์„œ ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜๋ฅผ ์ถ”์ •ํ•  ์ˆ˜ ์žˆ๋‹ค. ๊ธฐ์ค€ ์‹ ํ˜ธ์˜ TD๋Š” 150 $ns$, BW๋Š” 40 MHz, CF๋Š” 20 MHz๋กœ ์„ค์ •ํ•˜์˜€๊ณ , ์ค‘์‹ฌ ์‹œ๊ฐ„์„ 500 $ns$๋กœ ์„ค์ •ํ•˜์—ฌ ์‹คํ—˜์„ ์ง„ํ–‰ํ•˜์˜€๋‹ค. ๋กœ๋“œ ์ž„ํ”ผ๋˜์Šค์— ๋”ฐ๋ฅธ IF ์ธก์ • ์‹œ ์„ค๊ณ„๋˜๋Š” ๊ธฐ์ค€ ์‹ ํ˜ธ๋Š” ๋ชจ๋‘ ๋™์ผํ•˜๋‹ค. ์„ค๊ณ„ํ•œ ์‹œ๊ฐ„ ์˜์—ญ์—์„œ ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋Š” ๊ทธ๋ฆผ 3๊ณผ ๊ฐ™๋‹ค.

์ „ํŒŒ ์ƒ์ˆ˜๋Š” ์„ ๋กœ ์ •์ˆ˜์˜ ์˜ํ–ฅ์„ ๋ฐ›๊ธฐ ๋•Œ๋ฌธ์— ์ผ€์ด๋ธ” ์ข…๋‹จ์˜ ๋กœ๋“œ ์ž„ํ”ผ๋˜์Šค์— ๋”ฐ๋ผ ์ „ํŒŒ ์ƒ์ˆ˜๊ฐ€ ๋‹ฌ๋ผ์ง„๋‹ค. ์ผ€์ด๋ธ”์˜ ์—ดํ™”์ •๋„ ๋ฐ ๊ณ ์žฅ์ ์˜ ์œ ๋ฌด ๋“ฑ ์„ ๋กœ ์ •์ˆ˜์— ์˜ํ–ฅ์„ ์ฃผ๋Š” ๋‹ค์–‘ํ•œ ์š”์ธ์— ๋”ฐ๋ผ ์ „ํŒŒ ์ƒ์ˆ˜๋Š” ๋ณ€ํ™”ํ•œ๋‹ค. ๋ณธ ์‹คํ—˜์—์„œ๋Š” ๋กœ๋“œ ์ž„ํ”ผ๋˜์Šค๊ฐ€ ๊ฐœ๋ฐฉ ์ƒํƒœ์ผ ๋•Œ ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ์ง„ํ–‰ํ•˜์˜€๋‹ค. ์ „ํŒŒ ์ƒ์ˆ˜ ํŒŒ๋ผ๋ฏธํ„ฐ๋Š” ํ‘œ 1๊ณผ ๊ฐ™๋‹ค (18). ์ธ๊ฐ€ ์‹ ํ˜ธ์— ๋Œ€ํ•œ ์ „ํŒŒ ์‹ ํ˜ธ๋Š” 150 m, 300 m, 450 m, 600 m ๊ฐ„๊ฒฉ์œผ๋กœ ์ทจ๋“ํ–ˆ๋‹ค.

๊ทธ๋ฆผ. 2. Pxie๋ฅผ ์ด์šฉํ•œ ๋ฐ˜์‚ฌํŒŒ ๊ณ„์ธก๋ฒ• ์‹œ์Šคํ…œ

Fig. 2. Refletometry system using Pxie

../../Resources/kiee/KIEE.2023.72.6.757/fig2.png

๊ทธ๋ฆผ. 3. ์‹œ๊ฐ„ ์˜์—ญ์—์„œ ๊ธฐ์ค€ ์‹ ํ˜ธ์™€ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ

Fig. 3. Reference signal and Wigner-Ville Distribution

../../Resources/kiee/KIEE.2023.72.6.757/fig3.png

ํ‘œ 1. ์ „ํŒŒ ์ƒ์ˆ˜ ํŒŒ๋ผ๋ฏธํ„ฐ

Table 1. Propagation constant parameters

ํŒŒ๋ผ๋ฏธํ„ฐ

๊ฐ’

$\dot\alpha_{0}$

$1.1216\times 10^{-10}$

$\dot\beta_{0}$

$5.5690\times 10^{-9}$

$\ddot\alpha_{0}$

$-2.1787\times 10^{-19}$

$\ddot\beta_{0}$

$-2.0324\times 10^{-19}$

๊ทธ๋ฆผ. 4. ์‹œ๊ฐ„ ์˜์—ญ์—์„œ ์ „ํŒŒ ์‹ ํ˜ธ

Fig. 4. Propagation signal in time domain

../../Resources/kiee/KIEE.2023.72.6.757/fig4.png

์ธ๊ฐ€ ์‹ ํ˜ธ์˜ ์ „ํŒŒ์— ๋”ฐ๋ฅธ ์‹œ๊ฐ„ ์˜์—ญ์—์„œ์˜ ๋ณ€ํ™”๋Š” ๊ทธ๋ฆผ 4์™€ ๊ฐ™์œผ๋ฉฐ ์‹œ๊ฐ„ ์˜์—ญ์—์„œ ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ฆ๊ฐ€ํ•จ์— ๋”ฐ๋ผ ์‹ ํ˜ธ์˜ ๊ฐ์‡ ๊ฐ€ ์ปค์ ธ ์‹ ํ˜ธ์˜ ํ”ผํฌ ๊ฐ’์ด ๊ฐ์†Œํ•˜๋Š” ๊ฒƒ์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค.

์‹ ํ˜ธ๊ฐ€ ๋งค์šฐ ์ž‘์•„์ ธ ํŒŒํ˜•์œผ๋กœ๋Š” ๋ถ„์‚ฐ์„ ํ™•์ธํ•  ์ˆ˜ ์—†์–ด ์ „ํŒŒ ์ƒ์ˆ˜๋ฅผ ์ด๊ณ„๋„ํ•จ์ˆ˜๊นŒ์ง€ ๊ณ ๋ คํ•œ ์ฃผํŒŒ์ˆ˜ ์ข…์† ํ•„ํ„ฐ $G(z,\:w)= e^{-jk(w)z}$์™€ ํ‘œ 1์„ ์‚ฌ์šฉํ•˜์—ฌ ๊ฒฐ๊ณผ๋ฅผ ๋‚˜ํƒ€๋ƒˆ๋‹ค. ์ „ํŒŒ ๊ฑฐ๋ฆฌ์— ๋”ฐ๋ฅธ ์ฃผํŒŒ์ˆ˜ ๋ถ„์‚ฐ์„ ํ™•์ธํ•˜๊ธฐ ์œ„ํ•ด ๊ณ ์† ํ‘ธ๋ฆฌ์— ๋ณ€ํ™˜์„ ์ง„ํ–‰ํ•˜์—ฌ ๊ธฐ์ค€ ์‹ ํ˜ธ๋ฅผ ์ฃผํŒŒ์ˆ˜ ์˜์—ญ์—์„œ ํ™•์ธํ•˜์˜€๊ณ  ๊ฒฐ๊ณผ๋Š” ๊ทธ๋ฆผ 5์™€ ๊ฐ™๋‹ค. ๊ทธ๋ฆผ 5๋Š” ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ฆ๊ฐ€ํ• ์ˆ˜๋ก ๊ณ ์ฃผํŒŒ ์„ฑ๋ถ„์ด ์ €์ฃผํŒŒ ์„ฑ๋ถ„๋ณด๋‹ค ๊ฐ์†Œ๊ฐ€ ์‹ฌํ•˜๊ณ  ์ด๋กœ ์ธํ•ด ์ค‘์‹ฌ ์ฃผํŒŒ์ˆ˜๊ฐ€ ๋ณ€ํ™”๋ฅผ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค. ์ด๋กœ ์ธํ•ด ๊ฑฐ๋ฆฌ๊ฐ€ ์ „ํŒŒ๋˜๋Š” ๋™์•ˆ์— ์ „ํŒŒ ์†๋„์˜ ๋ณ€ํ™”๋ฅผ ์œ ์ถ”ํ•  ์ˆ˜ ์žˆ๋‹ค.

๋กœ๋“œ ์ž„ํ”ผ๋˜์Šค๊ฐ€ ๊ฐœ๋ฐฉ๋œ ์ƒํƒœ์ผ ๋•Œ ์ „ํŒŒ๋œ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋Š” ๊ทธ๋ฆผ 6๊ณผ ๊ฐ™๋‹ค. ๊ทธ๋ฆผ 6์—์„œ ์žฅ๊ฑฐ๋ฆฌ๋กœ ์ „ํŒŒ๋ ์ˆ˜๋ก ๊ฐ์‡ ๋ฅผ ๊ฒช์–ด 150 m ์ง€์  ์ดํ›„๋กœ๋Š” ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋ฅผ ํ†ตํ•ด ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜๋ฅผ ์ถ”์ •ํ•  ์ˆ˜ ์—†๋‹ค. ์ด์— ๋”ฐ๋ผ ๋ณธ ๋…ผ๋ฌธ์—์„œ ์ œ์‹œํ•œ ์‹ ํ˜ธ ๋ณต์› ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ์ ์šฉํ•˜์—ฌ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์˜ ํฌ๊ธฐ๋ฅผ ์ธ๊ฐ€ ์‹ ํ˜ธ์™€ ๋™์ผํ•˜๊ฒŒ ์กฐ์ •ํ•œ ํ›„ IF๋ฅผ ์ถ”์ •ํ•˜์˜€๋‹ค. IF ์ถ”์ • ์•Œ๊ณ ๋ฆฌ์ฆ˜์ด ์ ์šฉ๋œ ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋Š” ๊ทธ๋ฆผ 7๊ณผ ๊ฐ™๋‹ค.

์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์— ๋”ฐ๋ผ ์ธ๊ฐ€ ์‹ ํ˜ธ๊ฐ€ ์žฅ๊ฑฐ๋ฆฌ๋กœ ์ „ํŒŒ๋ ์ˆ˜๋ก ์ €์ฃผํŒŒ ๋ถ€๋ถ„๋ณด๋‹ค ๊ณ ์ฃผํŒŒ ๋ถ€๋ถ„์—์„œ ๋ถ„์‚ฐ์ด ํฌ๊ฒŒ ์ผ์–ด๋‚จ์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค.

๊ทธ๋ฆผ. 5. ์ „ํŒŒ ์‹ ํ˜ธ์˜ ์ฃผํŒŒ์ˆ˜ ์‘๋‹ต

Fig. 5. Frequency response of propagation signal

../../Resources/kiee/KIEE.2023.72.6.757/fig5.png

๊ทธ๋ฆผ. 6. ์ „ํŒŒ๋œ ๊ธฐ์ค€ ์‹ ํ˜ธ์˜ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ

Fig. 6. Wigner-Ville Distribution of propagated signal

../../Resources/kiee/KIEE.2023.72.6.757/fig6.png

๊ทธ๋ฆผ. 7. ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์—์„œ ์ถ”์ •ํ•œ IF

Fig. 7. IF estimation in Wigner-Vllie Distribution

../../Resources/kiee/KIEE.2023.72.6.757/fig7.png

๋กœ๋“œ ์ž„ํ”ผ๋˜์Šค๊ฐ€ ๊ฐœ๋ฐฉ๋œ ์ƒํƒœ์—์„œ ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ฆ๊ฐ€ํ• ์ˆ˜๋ก ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜์˜ ๊ธฐ์šธ๊ธฐ๊ฐ€ ๊ฐ์†Œํ•˜๊ฒŒ ๋˜๋Š”๋ฐ ์ด๋Š” ์ „ํŒŒ ์ƒ์ˆ˜์— ๋”ฐ๋ผ ์‹ ํ˜ธ๊ฐ€ ์ „ํŒŒ๋ ์ˆ˜๋ก ๊ณ ์ฃผํŒŒ์˜ ๋ถ„์‚ฐ์ด ์‹ฌํ•ด ์ €์ฃผํŒŒ๋ณด๋‹ค ๋„์ฐฉ ์‹œ๊ฐ„์ด ๋Šฆ์–ด์ ธ ๊ธฐ์šธ๊ธฐ๊ฐ€ ๊ฐ์†Œํ•˜๊ฒŒ ๋œ๋‹ค. ๋กœ๋“œ ์ž„ํ”ผ๋˜์Šค๊ฐ€ ๊ฐœ๋ฐฉ ์ƒํƒœ์ผ ๋•Œ ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜์˜ ๊ธฐ์šธ๊ธฐ๋Š” ํ‘œ 2๊ณผ ๊ฐ™๋‹ค.

ํ‘œ 2. ์ „ํŒŒ ๊ฑฐ๋ฆฌ์— ๋”ฐ๋ฅธ IF์˜ ๊ธฐ์šธ๊ธฐ

Table 2. Slope of IF along propagation distance

์ „ํŒŒ ๊ฑฐ๋ฆฌ

์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜์˜ ๊ธฐ์šธ๊ธฐ

0 [ $m$ ]

$1.061\times 10^{15}$

150 [ $m$ ]

$1.390\times 10^{15}$

300 [ $m$ ]

$2.924\times 10^{15}$

450 [ $m$ ]

$-3.455\times 10^{16}$

600 [ $m$]

$-5.326\times 10^{15}$

4. ๊ฒฐ ๋ก 

๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ์žฅ๊ฑฐ๋ฆฌ ์ „ํŒŒ์— ๋”ฐ๋ฅธ ์‹ ํ˜ธ์˜ ๊ฐ์‡„์™€ ๋ถ„์‚ฐ์ด ๊ณ ๋ ค๋œ ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜ ์ถ”์ • ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ๊ฐœ๋ฐœํ•˜์˜€์œผ๋ฉฐ ์ „ํŒŒ ์†๋„์˜ ๊ฒฝํ–ฅ์„ฑ์„ ํŒŒ์•…ํ•˜๊ณ  ์žฅ๊ฑฐ๋ฆฌ ์„ ๋กœ์— ์ ์šฉ ๊ฐ€๋Šฅํ•œ ์‹œ๋ฎฌ๋ ˆ์ดํ„ฐ๋ฅผ ๊ฐœ๋ฐœํ•˜์˜€๋‹ค. ๊ฐ์‡„๊ฐ€ ์ผ์–ด๋‚œ ์‹ ํ˜ธ์— ์‹ ํ˜ธ ๋ณต์› ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ์ ์šฉํ•˜์—ฌ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ์˜ ํฌ๊ธฐ๋ฅผ ์กฐ์ •ํ•œ ํ›„ ์œ„๊ทธ๋„ˆ-๋นŒ ๋ถ„ํฌ๋ฅผ ํ†ตํ•ด ์‹œ๊ฐ„-์ฃผํŒŒ์ˆ˜ ์˜์—ญ์—์„œ ์ˆœ์‹œ ์ฃผํŒŒ์ˆ˜๋ฅผ ์ถ”์ •ํ•˜์˜€๋‹ค. ์ „ํŒŒ ์†๋„๋Š” ์œ„์ƒ ์ƒ์ˆ˜์— ์˜ํ•ด ๋ณ€ํ™”ํ•˜๋ฉฐ ์ „ํŒŒ ๊ฑฐ๋ฆฌ๊ฐ€ ์ฆ๊ฐ€ํ• ์ˆ˜๋ก ๊ฐ์‡„ ๋ฐ ๋ถ„์‚ฐ์ด ์‹ฌํ•ด์ง€๋ฉฐ ๊ณ ์ฃผํŒŒ ์„ฑ๋ถ„์ด ์ €์ฃผํŒŒ๋ณด๋‹ค ๋งŽ์ด ์™œ๊ณก๋˜๋ฉฐ, ์ด์— ๋”ฐ๋ผ ์ฃผํŒŒ์ˆ˜ ์„ฑ๋ถ„ ๋ณ„ ์ „ํŒŒ ์†๋„๊ฐ€ ๋ณ€ํ™”ํ•œ๋‹ค. ์ด๋Š” ๊ณ ์žฅ์  ํƒ์ง€ ์„ฑ๋Šฅ์˜ ์ €ํ•˜๋กœ ์ด์–ด์ง€๋ฉฐ ์ด๋ฅผ ํ•ด๊ฒฐํ•˜๊ณ ์ž ์ฃผํŒŒ์ˆ˜์— ๋”ฐ๋ฅธ ์ „ํŒŒ ์†๋„๋ฅผ ๋ณด์ •ํ•˜๋Š” ์•Œ๊ณ ๋ฆฌ์ฆ˜์˜ ๊ฐœ๋ฐœ์ด ํ•„์š”ํ•˜๋‹ค. ๋˜ํ•œ ์‹ ํ˜ธ ๋ณต์› ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ๊ธฐ๋ฐ˜์œผ๋กœ ๋ณด์ƒ ํ•„ํ„ฐ์„ค๊ณ„๋ฅผ ํ•  ์˜ˆ์ •์ด๋‹ค.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) : 2021R1A2C1095779. This research was supported by Korea Electric Power Corporation.(Grant number: R22XO05-03)

References

1 
Chae-Kyun Jung, 2020 01, Development of HVDC Cable Diagnostic Technology, KIEE, World of Electricity, Vol. 69, No. 1, pp. 11-19Google Search
2 
Kuk Kim, Ik-soo Eo, 202001, Implementation of Power Cable Diagnostic Simulator using VLF, Journal of the Korea Academia-Industrial, Vol. 21, No. 8, pp. 593-602Google Search
3 
Jae-Seong Im, Keon-Hee Park, Sung-ho Noh, Woo-sung Jung, Jang-Seob Lim, 2020.11, Statisticla Compensation Scheme of VLF tanฮด Criteria for Application of Long Distance Cable, KIEE, KIEE Electrical Properties and Application Sector Autumn Conference, pp. 39-40Google Search
4 
Jae-Bong Lee, Yeon-Ha Jung, 201009, An Amendment of the VLF tanฮด Criteria to Improve the Diagnostic Accuracy of the XLPE-insulated Power Cables, KIEE, Vol. 59, No. 9, pp. 1644-1651Google Search
5 
Jong Min Jeong, Chun Ku Lee, Tae Sung Yoon, Jin Bae Park, 2014.12, Image Processing Based Time-Frequency Domain Reflectometry for Estimating the Fault Location Close to the Applied Signal Point, KIEE, Vol. 63, No. 12, pp. 1683-1689Google Search
6 
Chun-Kwon Lee, Seung Jin Chang, February 15 2021, A method of Fault Localization Within the Blind Spot Using the Hybridization Between TDR and Wavelet Transform, IEEE sens., Vol. 21, No. 4Google Search
7 
You Chung, Cynthia Furse, Jeremy Pruitt, May 2005, Application of Phase Detection Frequency Domain Reflectometry for Locating Faults in an F-18 Flight Control Harness, IEEE trans. Elect. Comp., Vol. 47, No. 2DOI
8 
Su Sik Bang, Seung-Jin Jang, Geon Seok Lee, Gu-Young Kwon, Yeong Ho Lee, Min-Jae Hwang Song-ho Sohn, Kijun Park, Yong-June Shin, 2015 07, EMTP Simulations and Analysis for Detection of Fault Location in High Temperature Superconducting Cables, KIEE The 46th KIEE Summer Conference 2015, pp. 357-358Google Search
9 
Chun Ku Lee, Jin Bae Park, Tae Sung Yoon, 201307, Instantaneous Frequency Estimation of the Gaussian Enveloped Linear Chirp Signal for Localizing the Faults of the Instrumental Cable in Nuclear Power Plant, KIEE, Vol. 62, No. 7, pp. 987-993Google Search
10 
Hyun-Mo Seong, Yeon-Sub Sim, Juin Go, Noah Yoo, Deokgyeom Kim, Hyeon Hong Seung Jin Chang, 2022 07, Development of a Reflectometry Simulator Considering Attenuation and Dispersion of Cable, KIEE The 53th KIEE Summer Conference 2022, pp. 274-275Google Search
11 
C. K. Lee, G. Y. Kwon, S. J. Chang, M. K. Jung, J. B. Park, H-S. Kim, Y. J. Shin, July 2017, Real-Time Condition Monitoring of LOCA via Time-Frequency Domain Reflectometry, IEEE trans Instrum Meas, Vol. 66, No. 7, pp. 1864-1873DOI
12 
J. Sophocles, 2002, Electromagnetic Wave and Antennas, Orfanidis, pp. 97-104Google Search
13 
Geon Seok Lee, Gu-Young Kwon, Su Sik Bang, Yeong Ho Lee, Song-Ho Sohn, Kijun Park, Yong-June Shin, June 2017, Monitoring Electrical and Thermal Characteristics of HTS Cable Systems via Timeโ€“Frequency Domain Reflectometry, IEEE trans Appl Supercon, Vol. 27, No. 4DOI
14 
S. J. Chang, Mar 2019, Wire Mismatch Detection Using a Convolutional Neural Network and Fault Localization Based on Time-Frequency-Domain Reflectometry, IEEE trans ind Electron, Vol. 66, No. 3, pp. 2102-2110DOI
15 
Yeon-Sub Sim, Hyun-Mo Seong, Seung Jin Chang, 2022 07, An Analysis of Propagation Velocity Tendency along the Propagation Distance of Cable, KIEE The 53th KIEE Summer Conference 2022, pp. 1813-1814Google Search
16 
Y.-J. Shin, Dec 2005, Application of time-frequency domain reflectometry for detection and localization of a fault on a coaxial cable, IEEE Trans Instrum Meas., Vol. 54, No. 6, pp. 2493-2500DOI
17 
B. Boashash, Apr 1992, Estimating and Interpreting the Instantaneous Frequency of a Signal-Part 2: Algorithms and Applications, Proc. IEEE, Vol. 80, pp. 540-568DOI
18 
C. K. Lee, G. Y. Kwon, Y. J. Shin, January 2019, Condition Assessment of I&C Cables in Nuclear Power Plants via Stepped-Frequency Waveform Reflectometry, IEEE trans. Instrum. Meas., Vol. 68, No. 1, pp. 215-224Google Search

์ €์ž์†Œ๊ฐœ

์„ฑํ˜„๋ชจ (Hyun-Mo Seong)
../../Resources/kiee/KIEE.2023.72.6.757/au1.png

he was born in Daejeon, South Korea.

He has been receiving the B.S. degrees from the Department of Electrical Engineering, Hanbat National University, Daejeon, South Korea, since 2018.

His general research interests include condition monitoring based on machine/ deep learning, diagnosis and prognostics of power equipment, signal processing techniques, and time-frequency analysis.

์‹ฌ์—ฐ์„ญ (Yeon-Sub Sim)
../../Resources/kiee/KIEE.2023.72.6.757/au2.png

He was born in Daejeon, South Korea.

He received the B.S. degrees from the Department of Electrical Engineering, Hanbat National University, Daejeon, South Korea, in 2022, where he is currently pursuing the M.S. degrees.

His general research interests include asset management systems based on machine/deep learning and efficient data analysis for complex data and the related applications.

์žฅ์Šน์ง„ (Seung Jin Chang)
../../Resources/kiee/KIEE.2023.72.6.757/au3.png

He was born in Seoul, South Korea.

He received the B.S. and Ph.D. degrees from the Department of Electrical and Electronic Engineering, Yonsei University, Seoul, in 2010 and 2017, respectively.

In 2018, he joined the School of Electrical and Computer Engineering, Seoul National University, Seoul, as a Post-Doctoral Researcher.

In 2018, he also joined the Department of Electrical Engineering, Hanbat National University, Daejeon South Korea, as an Assistant Professor, where he is currently an Associate Professor.

His current research interests are characterized by condition monitoring based on machine/deep learning, diagnosis and prognostics of power equipment, including cables and batteries, applied signal processing techniques, and timeโ€“frequency analysis.