• ๋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. (Agency for Defense Development)
  2. (Department of Electrical and Electronic Engineering, Hanyang University, Korea)



Target lock-on, Target tracking, Probability of target existence, Clutter

1. ์„œ ๋ก 

๋ ˆ์ด๋”(Radar) ์‹œ์Šคํ…œ์€ ์ „์žํŒŒ ์‹ ํ˜ธ๋ฅผ ๊ณต๊ฐ„์œผ๋กœ ๋ฐฉ์‚ฌํ•˜๊ณ  ๋Œ€์ƒ ํ‘œ์ ์œผ๋กœ๋ถ€ํ„ฐ ๋ฐ˜์‚ฌ๋˜๋Š” ์‹ ํ˜ธ์— ๋Œ€ํ•œ ์ˆ˜์‹  ๋ฐ ๋ถ„์„์„ ํ†ตํ•ด ํ‘œ์  ์ •๋ณด๋ฅผ ์ถ”์ถœํ•˜๋Š” ์„ผ์„œ์ด๋‹ค. ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ํƒ์ง€ ์„ฑ๋Šฅ์€ ์ผ๋ฐ˜์ ์œผ๋กœ ์‹œ์Šคํ…œ ์žก์Œ์„ ๊ณ ๋ คํ•œ ์ˆ˜์‹  ์‹ ํ˜ธ์˜ ์‹ ํ˜ธ ๋Œ€ ์žก์Œ๋น„(Signal-to-Noise Ratio ; SNR)์— ์˜ํ•ด ๊ฒฐ์ •๋˜๋Š”๋ฐ, SNR ๋“ฑ์„ ๊ณ ๋ คํ•˜์—ฌ ํ‘œ์ ์— ๋Œ€ํ•œ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ์ตœ๋Œ€ ํƒ์ง€๊ฑฐ๋ฆฌ๊ฐ€ ์ด๋ก ์ ์œผ๋กœ ์‚ฐ์ถœ๋˜๊ณ , ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์€ ์˜ˆ์ƒ๋˜๋Š” ์ตœ๋Œ€ ํƒ์ง€๊ฑฐ๋ฆฌ๋ฅผ ๊ธฐ์ค€์œผ๋กœ ํ‘œ์  ํƒ์ƒ‰์„ ์ˆ˜ํ–‰ํ•œ๋‹ค. ๊ทธ๋Ÿฐ๋ฐ ํด๋Ÿฌํ„ฐ(clutter) ํ™˜๊ฒฝ์—์„œ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์ด ์šด์šฉ๋˜๋Š” ๊ฒฝ์šฐ์—๋Š” ํƒ์ƒ‰ ์˜์—ญ์—์„œ ํ‘œ์  ์‹ ํ˜ธ์™€ ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ๊ฐ€ ๋™์‹œ์— ์ˆ˜์‹ ๋˜๋ฏ€๋กœ, ํ‘œ์  ์‹ ํ˜ธ์— ๋Œ€ํ•œ ํƒ์ง€ ์„ฑ๋Šฅ์€ SNR์ด ์•„๋‹Œ ์‹ ํ˜ธ ๋Œ€ ํด๋Ÿฌํ„ฐ๋น„(Signal-to- Clutter Ratio ; SCR)์— ์˜์กดํ•  ์ˆ˜ ์žˆ์œผ๋ฉฐ, ์‹ ํ˜ธ ํƒ์ง€ ๊ฒฐ๊ณผ๊ฐ€ ํ‘œ์ ์œผ๋กœ๋ถ€ํ„ฐ ๊ธฐ์ธ(origination)๋œ ๊ฒƒ์ธ์ง€์— ๋Œ€ํ•œ ๋ถˆํ™•์‹ค์„ฑ(uncertainty)์œผ๋กœ ์ธํ•ด ์ถ”๊ฐ€์ ์ธ ํ™•์ธ(confirmation) ๊ณผ์ •์„ ํ†ตํ•ด ํ‘œ์  ์‹ ํ˜ธ์— ๋Œ€ํ•œ ์‹ ๋ขฐ์„ฑ์„ ํ–ฅ์ƒ์‹œํ‚จ ํ›„, ํ‘œ์  ํฌ์ฐฉ(lock-on) ๋ฐ ์ถ”์  ๋‹จ๊ณ„๋กœ ์ „ํ™˜ํ•˜๋Š” ๊ฐœ๋…์œผ๋กœ ์šด์šฉํ•ด์•ผ ํ•œ๋‹ค. ๋˜ํ•œ ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ํ‘œ์ ์— ๋Œ€ํ•œ ์‚ฌ์ „ ์ •๋ณด์˜ ๋†’์€ ๋ถˆํ™•์‹ค์„ฑ์œผ๋กœ ์ธํ•ด ์ƒ๋Œ€์ ์œผ๋กœ ๋„“์€ ์˜์—ญ์„ ํƒ์ƒ‰ํ•ด์•ผ ๋˜๋Š” ๊ฒฝ์šฐ์—๋Š” ํƒ์ƒ‰ ์˜์—ญ์„ ๋ถ„ํ• ํ•˜๊ณ  ๋ถ„ํ• ๋œ ์˜์—ญ์— ๋Œ€ํ•˜์—ฌ ์ˆœ์ฐจ์ ์œผ๋กœ ํ‘œ์  ํƒ์ƒ‰์„ ์ˆ˜ํ–‰ํ•ด์•ผ ๋˜๋Š”๋ฐ, ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ํšจ์œจ์ ์ธ ์ž์› ๊ด€๋ฆฌ(resource management)๋ฅผ ์œ„ํ•œ ์ฃผ์š” ์„ค๊ณ„ ํŒŒ๋ผ๋ฏธํ„ฐ(parameter)๋Š” ๋ถ„ํ• ๋œ ํƒ์ƒ‰ ์˜์—ญ์— ๋Œ€ํ•œ ํƒ์ƒ‰ ์ง€์† ์‹œ๊ฐ„์ด ๋˜๋ฉฐ, ์ด๋ฅผ ์œ„ํ•ด์„œ๋Š” ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ์˜ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์— ๋Œ€ํ•œ ์˜ˆ์ธก ๊ธฐ๋ฒ•์ด ํ•„์š”ํ•˜๋‹ค.

ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์€ ์ˆ˜์‹  ์‹ ํ˜ธ ํƒ์ง€ ๊ฒฐ๊ณผ๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ํ•œ ํ™•์ธ ๊ธฐ๋ฒ•์— ์˜ํ•ด ๊ฒฐ์ •๋˜๋Š”๋ฐ, ์ผ๋ฐ˜์ ์œผ๋กœ ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ ์„ ์•Œ๊ณ  ์žˆ๋‹ค๋Š” ์ „์ œ ์กฐ๊ฑดํ•˜์— ์‹œ๊ฐ„์— ๋”ฐ๋ฅธ ๋‹ค์ˆ˜ ํƒ์ง€ ๊ฒฐ๊ณผ๋ฅผ ์ด์šฉํ•˜๋Š” ๋ˆ„์ (cumulative) ํƒ์ง€ ํ™•๋ฅ  ๊ฐœ๋…, ๊ด€์‹ฌ ์˜์—ญ์— ๋Œ€ํ•œ ํ‘œ์  ์กด์žฌ ํ™•๋ฅ (Probability of Target Existence ; PTE) ๊ฐœ๋… ๋“ฑ์„ ์ด์šฉํ•˜๋Š” ํ‘œ์  ํ™•์ธ ๊ธฐ๋ฒ•์ด ์žˆ๋‹ค. ๋ˆ„์  ํƒ์ง€ ํ™•๋ฅ  ๊ธฐ๋ฐ˜ ํ‘œ์  ํ™•์ธ ๊ธฐ๋ฒ•์ธ $m-of-n$ ํƒ์ง€ ๋ฐฉ๋ฒ•์€ $n$ ๋ฒˆ ์Šค์บ”(scan)ํ•˜์—ฌ $m$ ๋ฒˆ ์ด์ƒ ์‹ ํ˜ธ๊ฐ€ ํƒ์ง€๋˜๋Š” ๊ฒฝ์šฐ์—๋Š” ํ•ด๋‹น ์‹ ํ˜ธ๋ฅผ ํ‘œ์  ์‹ ํ˜ธ๋กœ ๊ฐ„์ฃผํ•˜์—ฌ ํฌ์ฐฉํ•˜๋Š” ๊ฐœ๋…์ด๊ณ , ๋‹จ์ผ ์Šค์บ”์— ๋Œ€ํ•œ ํ‘œ์  ํƒ์ง€ ํ™•๋ฅ ์„ ๊ธฐ์ค€์œผ๋กœ ๋‹ค์ˆ˜ ์Šค์บ”์— ๋Œ€ํ•œ ๋ˆ„์  ํƒ์ง€ ํ™•๋ฅ ์€ ์ดํ•ญ ์ •๋ฆฌ(binominal theorem)์— ์˜ํ•ด ์ด๋ก ์ ์œผ๋กœ ๊ณ„์‚ฐ๋œ๋‹ค(1)ยญ(3). ๊ทธ๋ฆฌ๊ณ  PTE๋Š” ๋‹ค์ˆ˜ ์ธก์ •์น˜(measurement)๊ฐ€ ํš๋“๋  ์ˆ˜ ์žˆ๋Š” ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ํ˜„์žฌ๊นŒ์ง€ ํš๋“๋œ ๋ชจ๋“  ์ธก์ • ์ •๋ณด๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ๊ด€์‹ฌ ์˜์—ญ์—์„œ์˜ ํ‘œ์  ์กด์žฌ ์—ฌ๋ถ€๋ฅผ ํ™•๋ฅ ์ ์œผ๋กœ ํŒ๋‹จํ•  ์ˆ˜ ์žˆ๋Š” ์ง€์ˆ˜(index)๋กœ์„œ, ํ‘œ์  ์‹ ํ˜ธ์˜ ์œ„์น˜ ๋ถ„ํฌ ํŠน์„ฑ์€ ํด๋Ÿฌํ„ฐ ๋“ฑ ํ‘œ์ ์ด ์•„๋‹Œ ๋ฌผ์ฒด์— ์˜ํ•œ ์‹ ํ˜ธ์˜ ์œ„์น˜ ๋ถ„ํฌ ํŠน์„ฑ๊ณผ๋Š” ๋‹ค๋ฅด๋‹ค๋Š” ๊ทผ๋ณธ์ ์ธ ์ฐจ์ด์ ์„ ์ด์šฉํ•˜๋Š” ๊ฐœ๋…์ด๋ฉฐ, ์ผ๋ฐ˜์ ์œผ๋กœ ์ถ”์  ํ•„ํ„ฐ์— ๋Œ€ํ•œ ํŠธ๋ž™(track) ๊ด€๋ฆฌ์˜ ๊ธฐ์ค€์œผ๋กœ ์ด์šฉ๋˜๊ณ  ์žˆ๋‹ค(4)ยญ(6). ๋˜ํ•œ PTE ๊ธฐ๋ฐ˜ ํ‘œ์  ํ™•์ธ ๊ธฐ๋ฒ•์€ ํŠน์ • ์‹œ์  ๊ธฐ์ค€์œผ๋กœ PTE๊ฐ€ ์›ํ•˜๋Š” ์ž„๊ณ„๊ฐ’(threshold)์— ๋„๋‹ฌํ•  ๋•Œ๊นŒ์ง€ ๊ฑธ๋ฆฌ๋Š” ์‹œ๊ฐ„์„ ์˜ˆ์ธกํ•˜๊ณ , ์˜ˆ์ธก๋œ ์‹œ์ ์—์„œ ์‚ฐ์ถœ๋˜๋Š” PTE๋ฅผ ์ด์šฉํ•˜์—ฌ ๊ด€์‹ฌ ์˜์—ญ์— ๋Œ€ํ•œ ํ‘œ์  ์กด์žฌ ์—ฌ๋ถ€๋ฅผ ํ™•์ธํ•˜๋ฉฐ, ํ‘œ์ ์ด ์กด์žฌํ•œ๋‹ค๊ณ  ํŒ๋‹จ๋˜๋Š” ๊ฒฝ์šฐ์—๋Š” ํš๋“๋˜๋Š” ์ธก์ •์น˜์— ํ‘œ์ ์— ์˜ํ•œ ์ธก์ •์น˜๊ฐ€ ํฌํ•จ๋˜์–ด ์žˆ๋‹ค๊ณ  ๊ฐ„์ฃผํ•˜์—ฌ ํ•ด๋‹น ํŠธ๋ž™์„ ์œ ์ง€ํ•œ๋‹ค(7)ยญ(8). ๊ทธ๋Ÿฐ๋ฐ, ์ด๋Ÿฌํ•œ ํ‘œ์  ํ™•์ธ ๊ธฐ๋ฒ•์€ ์‚ฌ์ „์— ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ ์„ ์•Œ๊ณ  ์žˆ์–ด์•ผ ํ•˜๋Š”๋ฐ, ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ๋Š” ํด๋Ÿฌํ„ฐ๋กœ๋ถ€ํ„ฐ ๋ฐ˜์‚ฌ๋˜๋Š” ์‹ ํ˜ธ์˜ ์ „๋ ฅ์ด ์‹œ๊ฐ„๊ณผ ๊ณต๊ฐ„์— ๋”ฐ๋ผ ํฌ๊ฒŒ ๋ณ€๊ฒฝ๋  ์ˆ˜ ์žˆ์œผ๋ฏ€๋กœ(9)ยญ(12), SCR์— ์˜ํ–ฅ์„ ๋ฐ›๋Š” ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ ์„ ์ •ํ™•ํžˆ ์•Œ ์ˆ˜ ์—†๋Š” ํ˜„์‹ค์ ์ธ ๋ฌธ์ œ์ ์ด ์žˆ๋‹ค. ๊ทธ๋ฆฌ๊ณ  ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ ์— ๋Œ€ํ•œ ๊ทผ๋ณธ์ ์ธ ๊ฐ€์ •์€ ๊ฒฐ๊ณผ์ ์œผ๋กœ ํ‘œ์  ํฌ์ฐฉ ์„ฑ๋Šฅ์— ๋Œ€ํ•œ ์˜ˆ์ธก ์˜ค์ฐจ๋ฅผ ์œ ๋ฐœํ•  ์ˆ˜ ์žˆ๋‹ค.

๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ์šด์šฉ๋˜๋Š” ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์— ๋Œ€ํ•˜์—ฌ ํƒ์ง€ ๊ฒฐ๊ณผ๋ฅผ ์ด์šฉํ•˜์—ฌ ์ถ”์ •ํ•˜๋Š” ํ‘œ์  SCR๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ PTE๊ฐ€ ์›ํ•˜๋Š” ์ž„๊ณ„๊ฐ’์— ๋„๋‹ฌํ•˜๋Š” ์†Œ์š” ์‹œ๊ฐ„์„ ์˜ˆ์ธกํ•˜๋Š” ๊ธฐ๋ฒ•์„ ์ œ์•ˆํ•˜๊ณ , ์ปดํ“จํ„ฐ ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ํ†ตํ•ด ์ œ์•ˆ๋œ ๊ธฐ๋ฒ•์˜ ์„ฑ๋Šฅ์„ ๋ณด์—ฌ์ค€๋‹ค.

2. ๊ธฐ์กด์˜ PTE ๊ธฐ๋ฐ˜ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์˜ˆ์ธก ๊ธฐ๋ฒ•

์ผ๋ฐ˜์ ์ธ PTE๋Š” ๊ด€์‹ฌ ์˜์—ญ์— ๋‹ค์ˆ˜ ์ธก์ •์น˜๊ฐ€ ์กด์žฌํ•  ๊ฒฝ์šฐ์— ํ‘œ์  ์‹ ํ˜ธ 1๊ฐœ์˜ ์œ„์น˜๋Š” ์‹œ์Šคํ…œ ์ธก์ • ์žก์Œ์— ์˜ํ•œ ๊ฐ€์šฐ์‹œ์•ˆ(Gaussian) ๋ถ„ํฌ ํŠน์„ฑ์„ ๋”ฐ๋ฅด์ง€๋งŒ, ํ‘œ์ ์ด ์•„๋‹Œ ๋ฌผ์ฒด์— ์˜ํ•œ ์‹ ํ˜ธ์˜ ์œ„์น˜๋Š” ๊ท ์ผ(uniform) ๋ถ„ํฌ ํŠน์„ฑ์„ ๋”ฐ๋ฅธ๋‹ค๊ณ  ๊ฐ€์ •ํ•œ๋‹ค. ๊ทธ๋ฆฌ๊ณ  ํš๋“๋œ ๋ชจ๋“  ์ธก์ •์น˜๊ฐ€ ํ‘œ์  ์‹ ํ˜ธ๋กœ๋ถ€ํ„ฐ ๊ธฐ์ธ๋  ์ˆ˜ ์žˆ๋‹ค๋Š” ์ „์ œ ์กฐ๊ฑดํ•˜์— ํŠน์ • ์ธก์ •์น˜๊ฐ€ ํ‘œ์  ์‹ ํ˜ธ์— ํ•ด๋‹นํ•˜๋ฉด ๋‚˜๋จธ์ง€ ์ธก์ •์น˜๋Š” ํ‘œ์ ์ด ์•„๋‹Œ ๋ฌผ์ฒด์— ์˜ํ•œ ์‹ ํ˜ธ๋ผ๋Š” ๋ชจ๋“  ๊ฒฝ์šฐ์˜ ์ˆ˜๋ฅผ ๊ณ ๋ คํ•˜์—ฌ ๊ด€์‹ฌ ์˜์—ญ์—์„œ์˜ ํ‘œ์  ์กด์žฌ ์—ฌ๋ถ€๋ฅผ ํ™•๋ฅ ์ ์œผ๋กœ ํŒ๋‹จํ•œ๋‹ค(4).

$k$ ์‹œ์ ์˜ ์‚ฌํ›„(posterior) PTE๋Š” ๋ฒ ์ด์ฆˆ ์ •๋ฆฌ(Bayesโ€™ theorem)์— ์˜ํ•ด ์‹ (1)๊ณผ ๊ฐ™์ด ํ‘œํ˜„๋˜๊ณ , $k$ ์‹œ์ ์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„(likelihood ratio)์™€ $k$ ์‹œ์ ์˜ ์‚ฌ์ „(prior) PTE์— ์˜ํ•ด ๊ฒฐ์ •๋˜๋Š” ๊ฒƒ์„ ์•Œ ์ˆ˜ ์žˆ๋‹ค.

(1)
\begin{align*} P(\chi_{k}| Z^{k})=\dfrac{p(z_{k},\: m_{k}|\chi_{k},\: Z^{k-1})P(\chi_{k}| Z^{k-1})}{p(z_{k},\: m_{k}| Z^{k-1})}\\ =\dfrac{\Lambda_{k}P(\chi_{k}| Z^{k-1})}{P(\overline{\chi}_{k}| Z^{k-1})+\Lambda_{k}P(\chi_{k}| Z^{k-1})} \end{align*}

์—ฌ๊ธฐ์„œ, $P(\chi_{k}| Z^{k})$๋Š”

$k$ ์‹œ์ ์˜ ์‚ฌํ›„ PTE, $\chi_{k}$๋Š” $k$ ์‹œ์ ์— ํ‘œ์ ์ด ์กด์žฌํ•˜๋Š” ์‚ฌ๊ฑด(event), $\overline{\chi}_{k}$๋Š” $k$ ์‹œ์ ์— ํ‘œ์ ์ด ์กด์žฌํ•˜์ง€ ์•Š๋Š” ์‚ฌ๊ฑด, $z_{k}$๋Š” $k$ ์‹œ์ ์— ํš๋“๋œ ์ธก์ •์น˜ ์ง‘ํ•ฉ, $m_{k}$๋Š” $k$ ์‹œ์ ์— ํš๋“๋œ ์ธก์ •์น˜ ๊ฐœ์ˆ˜, $Z^{k}$๋Š” $k$ ์‹œ์ ๊นŒ์ง€ ํš๋“๋œ ๋ชจ๋“  ์ธก์ •์น˜ ์ง‘ํ•ฉ, $\Lambda_{k}$๋Š” $k$ ์‹œ์ ์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„, $P(\chi_{k}| Z^{k-1})$๋Š” $k$ ์‹œ์ ์˜ ์‚ฌ์ „ PTE๋ฅผ ์˜๋ฏธํ•œ๋‹ค.

$k$ ์‹œ์ ์˜ ์ธก์ •๋น„ ์šฐ๋„๋น„๋Š” ์‹ (2)์™€ ๊ฐ™์ด $k$ ์‹œ์ ์— ํš๋“๋œ ์ธก์ •์น˜์— ๋Œ€ํ•˜์—ฌ ํ‘œ์ ์ด ์กด์žฌํ•˜์ง€ ์•Š๋Š” ์‚ฌ๊ฑด์ด ์ฃผ์–ด์กŒ์„ ๋•Œ์˜ ์šฐ๋„ ํ•จ์ˆซ๊ฐ’๊ณผ ํ‘œ์ ์ด ์กด์žฌํ•˜๋Š” ์‚ฌ๊ฑด์ด ์ฃผ์–ด์กŒ์„ ๋•Œ์˜ ์šฐ๋„ ํ•จ์ˆซ๊ฐ’์˜ ๋น„(ratio)๋กœ ํ‘œํ˜„๋˜๋Š”๋ฐ, PTE๋Š” ์ถ”์  ํ•„ํ„ฐ์˜ ์ธก์ •์น˜ ์˜ˆ์ธก ์ •๋ณด ๊ธฐ๋ฐ˜์œผ๋กœ ์šฐ๋„ ํ•จ์ˆ˜๋ฅผ ์‚ฌ์šฉํ•˜๋Š” ๊ฐœ๋…์ž„์„ ์•Œ ์ˆ˜ ์žˆ๋‹ค.

(2)
\begin{align*} \Lambda_{k}=\dfrac{p(z_{k},\: m_{k}|\chi_{k},\: Z^{k-1})}{p(z_{k},\: m_{k}|\overline{\chi}_{k},\: Z^{k-1})}\\ =\dfrac{\sum_{i=0}^{m_{k}}p(z_{k},\: m_{k},\:\chi_{k,\:i}|\chi_{k},\: Z^{k-1})}{p(z_{k},\: m_{k},\:\chi_{k,\:0}|\overline{\chi}_{k},\: Z^{k-1})}\\ =(1-P_{D}P_{G})+P_{D}\sum_{i=1}^{m_{k}}\dfrac{N(z_{k,\:i};\overline{z}_{k},\: S_{k})}{\hat\lambda_{k}} \end{align*}

์—ฌ๊ธฐ์„œ, $\chi_{k,\: i}$๋Š” $k$ ์‹œ์ ์˜ ์ธก์ •์น˜ ์ง‘ํ•ฉ์—์„œ $i$-๋ฒˆ์งธ ์ธก์ •์น˜๊ฐ€ ํ‘œ์  ์‹ ํ˜ธ์ธ ์‚ฌ๊ฑด, $\chi_{k,\: 0}$๋Š” $k$ ์‹œ์ ์˜ ๋ชจ๋“  ์ธก์ •์น˜๊ฐ€ ํ‘œ์ ์ด ์•„๋‹Œ ๋ฌผ์ฒด์— ์˜ํ•œ ์‹ ํ˜ธ์ธ ์‚ฌ๊ฑด, $P_{D}$๋Š” ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ , $P_{G}$๋Š” ํ‘œ์  ์‹ ํ˜ธ๊ฐ€ ์ธก์ • ๊ฒŒ์ดํŠธ(gate) ์˜์—ญ์— ์กด์žฌํ•  ํ™•๋ฅ , $\hat\lambda_{k}$๋Š” $k$ ์‹œ์ ์˜ ํด๋Ÿฌํ„ฐ ์ธก์ •์น˜ ๋ฐ€๋„ ์ถ”์ •์น˜, $z_{k,\: i}$๋Š” $k$ ์‹œ์ ์˜ ์ธก์ •์น˜ ์ง‘ํ•ฉ์—์„œ $i$-๋ฒˆ์งธ ์ธก์ •์น˜, $\overline{z}_{k}$๋Š” $k$ ์‹œ์ ์˜ ์ธก์ •์น˜ ์˜ˆ์ธก ์ •๋ณด, $S_{k}$๋Š” $k$ ์‹œ์ ์˜ ์ธก์ •์น˜ ๊ณต๋ถ„์‚ฐ ์ •๋ณด, $N(\bullet)$๋Š” ๊ฐ€์šฐ์‹œ์•ˆ ํ™•๋ฅ  ๋ฐ€๋„ ํ•จ์ˆ˜๋ฅผ ์˜๋ฏธํ•œ๋‹ค.

์žฌ๊ท€์ ์ธ(recursive) ํ˜•ํƒœ์˜ PTE๋กœ ํ‘œํ˜„ํ•˜๊ธฐ ์œ„ํ•ด์„œ๋Š” $k$ ์‹œ์ ์˜ ์‚ฌ์ „ PTE๋ฅผ ์œ„ํ•œ ์˜ˆ์ธก ๋ชจ๋ธ์ด ํ•„์š”ํ•œ๋ฐ, ์ผ๋ฐ˜์ ์œผ๋กœ $k$ ์‹œ์ ์˜ ํ‘œ์  ์กด์žฌ ์‚ฌ๊ฑด์€ $k-1$ ์‹œ์ ์˜ ํ‘œ์  ์กด์žฌ ์‚ฌ๊ฑด์—๋งŒ ์˜ํ–ฅ์„ ๋ฐ›๋Š”๋‹ค๋Š” ์ „์ œ ์กฐ๊ฑดํ•˜์— ์‹ (3)๊ณผ ๊ฐ™์ด ๋งˆ์ฝ”ํ”„ ์ฒด์ธ(Markov chain)-1 ํ™•๋ฅ  ์ฒœ์ด ํ–‰๋ ฌ๋กœ ํ‘œํ˜„๋œ๋‹ค.

(3)
\begin{align*} \left[\begin{aligned}P(\chi_{k}| Z^{k-1})\\P(\overline{\chi}_{k}| Z^{k-1})\end{aligned}\right]=\begin{bmatrix}\pi_{11}&\pi_{12}\\\pi_{21}&\pi_{22}\end{bmatrix}^{T}\left[\begin{aligned}P(\chi_{k-1}| Z^{k-1})\\P(\overline{\chi}_{k-1}| Z^{k-1})\end{aligned}\right] \end{align*}

์—ฌ๊ธฐ์„œ, $\pi_{ij}$๋Š” $k-1$ ์‹œ์ ์˜ $j$-๋ฒˆ์งธ ์‚ฌ๊ฑด์ด ์ฃผ์–ด์กŒ์„ ๋•Œ $k$ ์‹œ์ ์˜ $i$-๋ฒˆ์งธ ์‚ฌ๊ฑด์ด ๋ฐœ์ƒํ•  ์กฐ๊ฑด๋ถ€ ํ™•๋ฅ ์„ ์˜๋ฏธํ•œ๋‹ค.

๊ทธ๋Ÿฐ๋ฐ, ํ™•๋ฅ  ์ฒœ์ด ํ–‰๋ ฌ์˜ ์›์†Œ๋Š” ํ‘œ์ ์— ์˜ํ•ด ํ•ด๋‹น ์‚ฌ๊ฑด์ด ์œ ์ง€๋˜๋Š” ํ‰๊ท  ์‹œ๊ฐ„ ๋“ฑ์„ ๊ณ ๋ คํ•˜์—ฌ ์ƒ์ˆ˜(constant)๋กœ ์„ค์ •ํ•˜๋Š”๋ฐ, ์ผ๋ฐ˜์ ์œผ๋กœ ํ™•๋ฅ  ์ฒœ์ด ํ–‰๋ ฌ์€ ํ•ญ๋“ฑ(identity) ํ–‰๋ ฌ๋กœ ๊ทผ์‚ฌํ™”(approximation) ํ•  ์ˆ˜ ์žˆ์œผ๋ฏ€๋กœ, ์‹ (1)์˜ ์‚ฌํ›„ PTE๋Š” ์‹ (4)์™€ ๊ฐ™์ด ํ‘œํ˜„๋  ์ˆ˜ ์žˆ๋‹ค(7)ยญ(8).

(4)
$\dfrac{P(\chi_{k}| Z^{k})}{P(\overline{\chi}_{k}| Z^{k})}\approx(\Lambda_{k})\dfrac{P(\chi_{k-1}| Z^{k-1})}{P(\overline{\chi}_{k-1}| Z^{k-1})}$

๊ทธ๋Ÿฌ๋ฉด, $k$ ์‹œ์ ์„ ๊ธฐ์ค€์œผ๋กœ $k+n$ ์‹œ์ ์˜ ์‚ฌํ›„ PTE๋Š” ์‹ (5)์™€ ๊ฐ™์ด ํ‘œํ˜„๋œ๋‹ค.

(5)
$\dfrac{P(\chi_{k+n}| Z^{k+n})}{P(\overline{\chi}_{k+n}| Z^{k+n})}\approx\prod_{i=1}^{n}(\Lambda_{k+i})\dfrac{P(\chi_{k}| Z^{k})}{P(\overline{\chi}_{k}| Z^{k})}$

$k$ ์‹œ์ ์„ ๊ธฐ์ค€์œผ๋กœ ์‚ฌํ›„ PTE๊ฐ€ ์›ํ•˜๋Š” ์ž„๊ณ„๊ฐ’์— ๋„๋‹ฌํ•˜๊ธฐ๊นŒ์ง€ ์†Œ์š”๋˜๋Š” ์‹œ๊ฐ„์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’($n_{k}^{c}$)์€ $k+1$ ์‹œ์ ์—์„œ $k+n$ ์‹œ์ ๊นŒ์ง€์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„๊ฐ€ ๋™์ผํ•˜๋‹ค๋Š” ์ „์ œ ์กฐ๊ฑดํ•˜์— ์‹ (6)๊ณผ ๊ฐ™์ด ์œ ๋„๋œ๋‹ค(7)ยญ(8).

(6)
$n_{k}^{c}= E\{n\}\approx\dfrac{\ln\left\{\dfrac{T_{C}}{1-T_{C}}\right\}-\ln\left\{\dfrac{P(\chi_{k}| Z^{k})}{P(\overline{\chi}_{k}| Z^{k})}\right\}}{E\left\{\ln\left(\Lambda_{k+1}\right)|\chi_{k+1}\right\}}$

์—ฌ๊ธฐ์„œ, $E\{\bullet\}$๋Š” ๊ธฐ๋Œ€(expectation) ํ•จ์ˆ˜, $\ln\{\bullet\}$๋Š” ์ž์—ฐ ๋กœ๊ทธ(natural logarithm) ํ•จ์ˆ˜, $T_{C}$๋Š” ํ‘œ์  ํฌ์ฐฉ์„ ์„ ์–ธํ•˜๊ธฐ ์œ„ํ•œ PTE์˜ ์ž„๊ณ„๊ฐ’์„ ์˜๋ฏธํ•œ๋‹ค.

์‹ (6)์— ์˜ํ•ด ๊ณ„์‚ฐ๋˜๋Š” ์†Œ์š” ์‹œ๊ฐ„์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’์€ ๊ด€์‹ฌ ์˜์—ญ์— ํ‘œ์ ์ด ์กด์žฌํ•˜๊ณ  ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ์ง€์†์ ์œผ๋กœ ํš๋“๋œ๋‹ค๋Š” ์ „์ œ ์กฐ๊ฑดํ•˜์— ํ‘œ์ ์„ ํฌ์ฐฉํ•˜๊ธฐ๊นŒ์ง€ ์†Œ์š”๋˜๋Š” ์‹œ๊ฐ„์„ ์˜๋ฏธํ•˜๋ฏ€๋กœ, ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์€ ํƒ์ƒ‰ ๋‹จ๊ณ„์—์„œ ํŠน์ • ์˜์—ญ์— ๋Œ€ํ•œ ํƒ์ƒ‰ ์ง€์† ์‹œ๊ฐ„์„ ํ•ด๋‹น ๊ธฐ๋Œ“๊ฐ’์„ ๊ธฐ์ค€์œผ๋กœ ์„ค๊ณ„ํ•˜๋ฉด ๋œ๋‹ค.

3. ์ œ์•ˆํ•˜๋Š” PTE ๊ธฐ๋ฐ˜ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์˜ˆ์ธก ๊ธฐ๋ฒ•

ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ์ด๋™ํ˜• ํ”Œ๋žซํผ(platform)์— ํƒ‘์žฌ๋˜์–ด ์žˆ๋Š” ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์ด ๋Œ€์ƒ ํ‘œ์ ์— ๋Œ€ํ•˜์—ฌ ์˜ˆ์ƒ๋˜๋Š” ์ตœ๋Œ€ ํƒ์ง€๊ฑฐ๋ฆฌ๋ถ€ํ„ฐ ํƒ์ƒ‰์„ ์ˆ˜ํ–‰ํ•˜๋Š” ๊ฒฝ์šฐ์—๋Š” ์‹ค์ œ ํƒ์ง€ ์„ฑ๋Šฅ์ด ํด๋Ÿฌํ„ฐ ๋ฐ˜์‚ฌ๊ณ„์ˆ˜ ํŠน์„ฑ ๋“ฑ์— ๋”ฐ๋ผ ๋ณ€๊ฒฝ๋  ์ˆ˜ ์žˆ๋Š” SCR์— ์˜์กดํ•˜๊ฒŒ ๋˜๋ฏ€๋กœ, ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ ์„ ์ •ํ™•ํžˆ ์•Œ ์ˆ˜ ์—†๋Š” ์ œํ•œ์ ์ด ์žˆ๋‹ค. ๊ทธ๋Ÿฐ๋ฐ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ์ดˆ๊ธฐ ํƒ์ƒ‰ ๋‹จ๊ณ„์—์„œ๋Š” ๊ทธ๋ฆผ 1์—์„œ ๋ณด๋Š” ๋ฐ”์™€ ๊ฐ™์ด ์‹ค์ œ ํƒ์ง€ ํ™•๋ฅ ($P_{D,\:P}$)์ด ์›ํ•˜๋Š” ํƒ์ง€ ํ™•๋ฅ ($P_{D,\:D}$)๋งŒํผ ์ถฉ์กฑ๋˜์ง€ ์•Š์„ ์ˆ˜ ์žˆ์œผ๋‚˜, ์šด์šฉ ์‹œ๊ฐ„์ด ์ง€๋‚จ์— ๋”ฐ๋ผ ํ‘œ์ ๊ณผ์˜ ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ๊ฐ€ ์ค„์–ด๋“ค๋ฉด์„œ SCR๊ณผ ์‹ค์ œ ํƒ์ง€ ํ™•๋ฅ ์ด ๋™์‹œ์— ์ฆ๊ฐ€ํ•˜๊ฒŒ ๋˜์–ด ๊ฒฐ๊ตญ์—๋Š” ์›ํ•˜๋Š” ํƒ์ง€ ํ™•๋ฅ ์— ๋„๋‹ฌํ•˜๊ฒŒ ๋  ๊ฒƒ์ด๋‹ค. ๋”ฐ๋ผ์„œ ์ด๋Ÿฌํ•œ ํ‘œ์  ์กฐ์šฐ ํ™˜๊ฒฝ์—์„œ ํƒ์ƒ‰ ์˜์—ญ์— ํ‘œ์ ์€ ์กด์žฌํ•˜์ง€๋งŒ ์šด์šฉ ์‹œ๊ฐ„์— ๋”ฐ๋ผ SCR์ด ๋ณ€๊ฒฝ๋˜๋Š” ์กฐ๊ฑด์„ ๊ณ ๋ คํ•œ PTE ๊ธฐ๋ฐ˜ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์˜ˆ์ธก ๊ธฐ๋ฒ•์„ ์ œ์•ˆํ•œ๋‹ค.

๊ทธ๋ฆผ 1 ์ œ์•ˆํ•˜๋Š” ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์˜ˆ์ธก ๊ธฐ๋ฒ•์˜ ์ „์ œ ์กฐ๊ฑด

Fig. 1 Condition of proposed method for predicting target lock-on time

../../Resources/kiee/KIEE.2022.71.4.663/fig1.png

์ผ๋ฐ˜์ ์œผ๋กœ ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ ์ด ๋„ˆ๋ฌด ๋‚ฎ์œผ๋ฉด PTE ์„ฑ๋Šฅ์ด ๋ณด์žฅ๋˜์ง€ ์•Š์„ ์ˆ˜ ์žˆ์œผ๋ฏ€๋กœ(6), ๊ทธ๋ฆผ 1์—์„œ ๋ณด๋Š” ๋ฐ”์™€ ๊ฐ™์ด ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ํƒ์ƒ‰ ์˜์—ญ์— ํ‘œ์ ์ด ์กด์žฌํ•˜๋Š”๋ฐ $P_{D,\:P}$๊ฐ€ $P_{D,\:D}$๋ณด๋‹ค ์ž‘์œผ๋ฉด ๊ฐ€์ •ํ•œ ํƒ์ง€ ํ™•๋ฅ ($P_{D,\:A}$)์ด 0์ธ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜์ง€ ์•Š๋Š” ์‚ฌ๊ฑด($\chi^{n}$)์œผ๋กœ ๊ฐ„์ฃผํ•˜๊ณ , $P_{D,\:P}$๊ฐ€ $P_{D,\:D}$๋ณด๋‹ค ํฌ๋ฉด $P_{D,\:A}$๊ฐ€ $P_{D,\:D}$์™€ ๋™์ผํ•˜๋ฉด์„œ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜๋Š” ์‚ฌ๊ฑด($\chi^{v}$)์œผ๋กœ ๊ฐ€์ •ํ•  ์ˆ˜ ์žˆ๋‹ค. ๋”ฐ๋ผ์„œ ์ดˆ๊ธฐ ํƒ์ƒ‰ ๋‹จ๊ณ„์—์„œ๋Š” ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์— ๋”ฐ๋ผ ์ด๋Ÿฌํ•œ ๋ชจ๋“  ์‚ฌ๊ฑด์ด ๋ฐœ์ƒ ๊ฐ€๋Šฅํ•˜๋ฏ€๋กœ, ๊ธฐ์กด ๊ธฐ๋ฒ•๊ณผ๋Š” ๋‹ฌ๋ฆฌ PTE ์˜ˆ์ธก ๋ชจ๋ธ์„ ๋งˆ์ฝ”ํ”„ ์ฒด์ธ-2 ํ™•๋ฅ  ์ฒœ์ด ํ–‰๋ ฌ๋กœ ์ •์˜ํ•˜์—ฌ์•ผ ํ•œ๋‹ค. ๋งˆ์ฝ”ํ”„ ์ฒด์ธ-2 ๋ชจ๋ธ์€ ์‹ (7)๊ณผ ๊ฐ™์ด ํ‘œ์ ์ด ์กด์žฌํ•˜๋ฉด์„œ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜์ง€ ์•Š๋Š” ์‚ฌ๊ฑด๊ณผ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜๋Š” ์‚ฌ๊ฑด์„ ๋™์‹œ์— ๊ณ ๋ คํ•œ๋‹ค(13)ยญ(15).

(7)
$$\left[\begin{array}{l}P\left(\chi_{k}^{v} \mid Z^{k-1}\right) \\ P\left(\chi_{k}^{n} \mid Z^{k-1}\right) \\ P\left(\overline{\chi_{k}} \mid Z^{k-1}\right)\end{array}\right]=\left[\begin{array}{lll}\pi_{11} \pi_{12} & \pi_{13} \\ \pi_{21} & \pi_{22} & \pi_{23} \\ \pi_{31} & \pi_{32} & \pi_{33}\end{array}\right]^{T}\left[\begin{array}{l}P\left(\chi_{k-1}^{v} \mid Z^{k-1}\right) \\ P\left(\chi_{k-1}^{n} \mid Z^{k-1}\right) \\ P\left(\overline{\chi_{k-1}} \mid Z^{k-1}\right)\end{array}\right]$$ $$\pi_{i j}=P(j \mid i), \quad \sum_{j=1}^{3} \pi_{i, j}=1$$ for $i=1$($\chi_{k-1}^{v}$),$2(\chi_{k-1}^{n}$),$3(\overline{\chi}_{k-1}$) and $j=1$($\chi_{k}^{v}$),$2(\chi_{k}^{n}$),$3(\overline{\chi}_{k}$)

์—ฌ๊ธฐ์„œ, $\chi_{k}^{v}$๋Š” $k$ ์‹œ์ ์— ํ‘œ์ ์ด ์กด์žฌํ•˜๋ฉด์„œ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜๋Š” ์‚ฌ๊ฑด, $\chi_{k}^{n}$๋Š” $k$ ์‹œ์ ์— ํ‘œ์ ์ด ์กด์žฌํ•˜์ง€๋งŒ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜์ง€ ์•Š๋Š” ์‚ฌ๊ฑด์„ ์˜๋ฏธํ•œ๋‹ค.

๋งˆ์ฝ”ํ”„ ์ฒด์ธ-2 ๋ชจ๋ธ ๊ธฐ๋ฐ˜ ์‚ฌํ›„ PTE๋Š” ์‹ (8)๊ณผ ๊ฐ™์ด ์‹ (2)์™€ ๋™์ผํ•œ ์ธก์ •์น˜ ์šฐ๋„๋น„๋ฅผ ์ด์šฉํ•˜๋Š” ํ™•๋ฅ ๊ณผ ์‹ (2)์—์„œ $P_{D}=0$์ธ ์ธก์ •์น˜ ์šฐ๋„๋น„๋ฅผ ์ด์šฉํ•˜๋Š” ํ™•๋ฅ ์˜ ํ•ฉ์œผ๋กœ ํ‘œํ˜„๋œ๋‹ค.

(8)
\begin{align*} P(\chi_{k}| Z^{k})=P(\chi_{k}^{n}| Z^{k})+P(\chi_{k}^{v}| Z^{k})\\ =\dfrac{P(\chi_{k}^{n}| Z^{k-1})+\Lambda_{k}P(\chi_{k}^{v}| Z^{k-1})}{1-(1-\Lambda_{k})P(\chi_{k}^{v}| Z^{k-1})} \end{align*}

์‹ (4)์™€ ๋™์ผํ•˜๊ฒŒ ๋งˆ์ฝ”ํ”„ ์ฒด์ธ-2 ํ™•๋ฅ  ์ฒœ์ด ํ–‰๋ ฌ์„ ํ•ญ๋“ฑ ํ–‰๋ ฌ๋กœ ๊ทผ์‚ฌํ™”ํ•˜๋ฉด, PTE ๊ธฐ๋ฐ˜ ์ถ”์  ํ•„ํ„ฐ์˜ ์ดˆ๊ธฐํ™” ์‹œ์ ์„ ๊ธฐ์ค€์œผ๋กœ $k$ ์‹œ์ ์˜ ์‚ฌํ›„ PTE๋Š” ์‹ (9)์™€ ๊ฐ™์ด ํ‘œํ˜„๋œ๋‹ค(7)ยญ(8).

(9)
\begin{align*} \dfrac{P(\chi_{k}| Z^{k})}{P(\overline{\chi}_{k}| Z^{k})}=(\Lambda_{k})\dfrac{P(\chi_{k-1}^{v}| Z^{k-1})}{P(\overline{\chi}_{k-1}| Z^{k-1})}+\dfrac{P(\chi_{k-1}^{n}| Z^{k-1})}{P(\overline{\chi}_{k-1}| Z^{k-1})}\\ \approx(\Lambda_{k}\Lambda_{k-1}\cdots\Lambda_{1})\dfrac{P(\chi_{0}^{v}| Z^{0})}{P(\overline{\chi}_{0}| Z^{0})}+\dfrac{P(\chi_{0}^{n}| Z^{0})}{P(\overline{\chi}_{0}| Z^{0})} \end{align*}

$k$ ์‹œ์ ์˜ ๋งˆ์ฝ”ํ”„ ์ฒด์ธ-2 ๋ชจ๋ธ ๊ธฐ๋ฐ˜ ์‚ฌํ›„ PTE๊ฐ€ ์›ํ•˜๋Š” ์ž„๊ณ„๊ฐ’($T_{C}$)๊ณผ ์ผ์น˜ํ•˜์˜€์„ ๋•Œ, ์‹ (9)๋Š” ์‹ (10)๊ณผ ๊ฐ™์ด ํ‘œํ˜„๋˜๊ณ , $k$ ์‹œ์ ๊นŒ์ง€์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„ ๊ณฑ์€ ์ƒ์ˆ˜๋กœ ๊ทผ์‚ฌํ™”๋  ์ˆ˜ ์žˆ๋‹ค.

(10)
$\prod_{i=1}^{k}\Lambda_{i}\approx\dfrac{P(\overline{\chi}_{0}|Z^{0})}{P(\chi_{0}^{v}|Z^{0})}\left\{\dfrac{T_{C}}{1-T_{C}}-\dfrac{P(\chi_{0}^{n}|Z^{0})}{P(\overline{\chi}_{0}|Z^{0})}\right\}=C_{tc}$

๊ทธ๋Ÿฐ๋ฐ, ๊ฐ ์‹œ์ ์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„๋Š” ์šด์šฉ ํ™˜๊ฒฝ์— ๋”ฐ๋ผ ์‹ (11)๊ณผ ๊ฐ™์ด ํƒ์ƒ‰ ์˜์—ญ์— ํ‘œ์ ์ด ์กด์žฌํ•˜์ง€๋งŒ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜์ง€ ์•Š๋Š” ์‚ฌ๊ฑด์ด ์ฃผ์–ด์กŒ์„ ๋•Œ์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’, ํ‘œ์ ์ด ์กด์žฌํ•˜๊ณ  ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜๋Š” ์ƒํ™ฉ์—์„œ ํ‘œ์  ์‹ ํ˜ธ๊ฐ€ ํƒ์ง€๋˜์ง€ ์•Š๋Š” ์‚ฌ๊ฑด์ด ์ฃผ์–ด์กŒ์„ ๋•Œ์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’, ๋˜๋Š” ํ‘œ์ ์ด ์กด์žฌํ•˜๊ณ  ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜๋Š” ์ƒํ™ฉ์—์„œ ํ‘œ์  ์‹ ํ˜ธ๊ฐ€ ํƒ์ง€๋˜๋Š” ์‚ฌ๊ฑด์ด ์ฃผ์–ด์กŒ์„ ๋•Œ์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’ ์ค‘ ํ•˜๋‚˜์˜ ๊ฐ’์œผ๋กœ ํ‘œํ˜„๋  ์ˆ˜ ์žˆ๋‹ค.

(11)
\begin{align*} \Lambda^{\alpha}= E\left\{\Lambda_{i}|\chi_{i}^{n}\right\}\\ \Lambda^{\beta}= E\left\{\Lambda_{i}|\chi_{i,\:0}^{v},\:\chi_{i}^{v}\right\}{for} i=1,\: ... ,\: k\\ \Lambda^{\gamma}= E\left\{\Lambda_{i}|\overline{\chi}_{i,\:0}^{v},\:\chi_{i}^{v}\right\} \end{align*}

์—ฌ๊ธฐ์„œ, $\chi_{i,\:0}^{v}$๋Š” $i$ ์‹œ์ ์— ํ‘œ์ ์ด ์กด์žฌํ•˜์ง€๋งŒ ๋ชจ๋“  ์ธก์ •์น˜๊ฐ€ ํ‘œ์ ์ด ์•„๋‹Œ ๋ฌผ์ฒด์— ์˜ํ•œ ์‚ฌ๊ฑด, $\overline{\chi}_{i,\:0}^{v}$๋Š” $i$ ์‹œ์ ์— ํ‘œ์ ์ด ์กด์žฌํ•˜๋ฉด์„œ ํ‘œ์ ์— ์˜ํ•œ ์ธก์ •์น˜๊ฐ€ ์กด์žฌํ•˜๋Š” ์‚ฌ๊ฑด์„ ์˜๋ฏธํ•œ๋‹ค.

๋”ฐ๋ผ์„œ, $k$ ์‹œ์ ๊นŒ์ง€์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„ ๊ณฑ์€ ์‹ (11)์˜ 3๊ฐ€์ง€ ์‚ฌ๊ฑด์— ๋Œ€ํ•œ ์ธก์ •์น˜ ์šฐ๋„๋น„์˜ ๊ธฐ๋Œ“๊ฐ’์œผ๋กœ ๋Œ€ํ‘œํ•˜์—ฌ ์‹ (12)์™€ ๊ฐ™์ด ํ‘œํ˜„๋  ์ˆ˜ ์žˆ๋‹ค.

(12)
$[\Lambda^{\alpha}]^{m}[\Lambda^{\beta}]^{n}[\Lambda^{\gamma}]^{l}\approx C_{tc},\: k =m+s ,\: s=n+l$

์‹ (12)์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„๋ฅผ ์ž์—ฐ ๋กœ๊ทธ ์Šค์ผ€์ผ๋กœ ๋ณ€ํ™˜ํ•œ ํ›„, ๊ธฐ๋Œ“๊ฐ’์œผ๋กœ ํ‘œํ˜„ํ•˜๋ฉด ์‹ (13)๊ณผ ๊ฐ™๋‹ค.

(13)
\begin{align*} E\left\{m |\chi^{n}\right\}\ln(\Lambda^{\alpha})+E\left\{n |\chi_{0}^{v},\:\chi^{v}\right\}\ln(\Lambda^{\beta})+\\ E\left\{l |\overline{\chi}_{0}^{v},\:\chi^{v}\right\}\ln(\Lambda^{\gamma})\approx\ln(C_{tc}) \end{align*}

$k$ ์‹œ์ ๊ณผ $n$ ์‹œ์ ์˜ ๊ธฐ๋Œ“๊ฐ’์€ ์‹ (12)์—์„œ ์ •์˜ํ•˜๊ณ  ์žˆ๋Š” ๊ฐ ์‚ฌ๊ฑด์˜ ์‹œ๊ฐ„ ๊ตฌ๊ฐ„์— ๋Œ€ํ•œ ๊ด€๊ณ„์— ์˜ํ•˜์—ฌ ์‹ (14)์™€ ๊ฐ™์ด ํ‘œํ˜„๋  ์ˆ˜ ์žˆ๋‹ค.

(14)
\begin{align*} E\{k |\chi\}= E\left\{m |\chi^{n}\right\}+E\left\{s |\chi^{v}\right\}\\ E\left\{n |\chi_{0}^{v},\:\chi^{v}\right\}=E\left\{s |\chi^{v}\right\}- E\left\{l |\overline{\chi}_{0}^{v},\:\chi^{v}\right\} \end{align*}

๊ทธ๋ฆฌ๊ณ , $ELL$ ์‹œ์ ์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’์€ ํ‘œ์ ์ด ์กด์žฌํ•˜๋ฉด์„œ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜๋Š” ์‚ฌ๊ฑด์ด ์ด $s$ ๋ฒˆ ์ค‘ $ELL$ ๋ฒˆ ๋ฐœ์ƒ๋˜๋Š” ์ดํ•ญ ์ •๋ฆฌ๋ฅผ ์ด์šฉํ•˜์—ฌ ์‹ (15)์™€ ๊ฐ™์ด ํ‘œํ˜„๋  ์ˆ˜ ์žˆ๋‹ค.

(15)
$E\left\{l |\overline{\chi}_{0}^{v},\:\chi^{v}\right\}=E\left\{s |\chi^{v}\right\}P_{D}P_{G}$

์ตœ์ข…์ ์œผ๋กœ ์‚ฌํ›„ PTE๊ฐ€ ์›ํ•˜๋Š” ์ž„๊ณ„๊ฐ’์— ๋„๋‹ฌํ•˜๋Š” $k$ ์‹œ์ ์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’์€ ์‹ (14)์™€ ์‹ (15)๋ฅผ ์ด์šฉํ•˜์—ฌ ์‹ (16)๊ณผ ๊ฐ™์ด ํ‘œํ˜„๋˜๊ณ , ์ด๋Š” ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์˜ ์˜ˆ์ธก๊ฐ’์— ํ•ด๋‹น๋œ๋‹ค.

(16)
\begin{align*} k_{0}^{c}=E[k |\chi]\approx E[m |\chi^{n}]+\\ \dfrac{\ln(C_{tc})}{(1-P_{D}P_{G})\ln(\Lambda^{\beta})+P_{D}P_{G}\ln(\Lambda^{\gamma})} \end{align*}

์‹ (16)์—์„œ ์ž์—ฐ ๋กœ๊ทธ ์Šค์ผ€์ผ์˜ ์ธก์ •์น˜ ์šฐ๋„๋น„๋Š” ์‹œ๋ฎฌ๋ ˆ์ด์…˜์— ์˜ํ•œ ๋ฐฉ๋ฒ• ๋˜๋Š” ์ด๋ก ์ ์ธ ํ•ด์„ ๋ฐฉ๋ฒ•์„ ์ด์šฉํ•˜์—ฌ ์‚ฐ์ถœํ•  ์ˆ˜ ์žˆ๋‹ค(8).

์ผ๋ฐ˜์ ์œผ๋กœ ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ์˜ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์€ ๊ธฐํ•˜ํ•™์ ์ธ ์กฐ๊ฑด์— ๋”ฐ๋ผ ํŽ„์Šค ์ œํ•œ์ (pulse-limited) ํ™˜๊ฒฝ๊ณผ ๋น” ์ œํ•œ์ (beam-limited) ํ™˜๊ฒฝ์œผ๋กœ ๊ตฌ๋ถ„๋œ๋‹ค. ๊ทธ๋ฆฌ๊ณ  SCR์€ ํ‘œ์  RCS(Radar Cross Section)์™€ ํด๋Ÿฌํ„ฐ RCS์˜ ๋น„(ratio)๋กœ ํ‘œํ˜„๋˜๊ณ , ํด๋Ÿฌํ„ฐ RCS๋Š” ํด๋Ÿฌํ„ฐ ๋ฐ˜์‚ฌ๊ณ„์ˆ˜์™€ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์ด ์ˆ˜์‹  ์‹ ํ˜ธ๋ฅผ ์ฒ˜๋ฆฌํ•˜๋Š” ํด๋Ÿฌํ„ฐ ๋‹จ๋ฉด์ ์— ์˜ํ•ด ๊ฒฐ์ •๋œ๋‹ค. ํด๋Ÿฌํ„ฐ ๋‹จ๋ฉด์ ์€ ํŽ„์Šค ์ œํ•œ์  ํ™˜๊ฒฝ์—์„œ๋Š” ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ์— ๋น„๋ก€ํ•˜๊ณ  ๋น” ์ œํ•œ์  ํ™˜๊ฒฝ์—์„œ๋Š” ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ์˜ ์ œ๊ณฑ์— ๋น„๋ก€ํ•˜๋ฏ€๋กœ, ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ SCR์€ ์‹ (17)๊ณผ ๊ฐ™์ด ํ‘œํ˜„๋œ๋‹ค(2)ยญ(3).

(17)
$SCR =\dfrac{RCS_{t}}{\sigma_{c}A_{c}}\propto\begin{cases} \dfrac{1}{R_{r}}&(pulse-li mi ted)\\ \dfrac{1}{R_{r}^{2}}&(beam-li mi ted) \end{cases}$

์—ฌ๊ธฐ์„œ, $RCS_{t}$๋Š” ํ‘œ์  RCS, $\sigma_{c}$๋Š” ํด๋Ÿฌํ„ฐ ๋ฐ˜์‚ฌ๊ณ„์ˆ˜, $A_{c}$๋Š” ํด๋Ÿฌํ„ฐ ๋‹จ๋ฉด์ , $R_{r}$์€ ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ๋ฅผ ์˜๋ฏธํ•œ๋‹ค.

๊ทธ๋ฆฌ๊ณ , ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์—๋Š” ํ‘œ์  ์‹ ํ˜ธ์™€ ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ๊ฐ€ ํ˜ผ์žฌ๋˜์–ด ์ˆ˜์‹ ๋œ๋‹ค. ํ‘œ์  ์‹ ํ˜ธ ์ „๋ ฅ๊ณผ ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ ์ „๋ ฅ์„ ๊ตฌ๋ถ„ํ•˜์—ฌ ์ธก์ •ํ•˜๊ธฐ ์–ด๋ ค์šฐ๋ฏ€๋กœ, SCR์€ ์ผ๋ฐ˜์ ์œผ๋กœ ์ •ํ™•ํžˆ ์•Œ ์ˆ˜ ์—†๋‹ค. ๊ทธ๋Ÿฐ๋ฐ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์—์„œ ์˜ค๊ฒฝ๋ณด(false alarm) ํ™•๋ฅ ์„ ๊ด€๋ฆฌํ•˜๊ธฐ ์œ„ํ•˜์—ฌ ๊ธฐ๋ณธ์ ์œผ๋กœ ์‚ฌ์šฉํ•˜๋Š” CFAR(Constant False Alarm Rate) ํƒ์ง€ ๊ธฐ๋ฒ•์€ ๋ฐฐ๊ฒฝ ์‹ ํ˜ธ์˜ ํ™•๋ฅ  ๋ฐ€๋„ ํ•จ์ˆ˜์— ๋Œ€ํ•œ ํ†ต๊ณ„์ ์ธ ๋ถ„์„ ๊ฒฐ๊ณผ๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ํƒ์ง€ ์ž„๊ณ„๊ฐ’์„ ์„ค์ •ํ•œ๋‹ค(1)ยญ(3). ๋”ฐ๋ผ์„œ ์ถ”์  ํ•„ํ„ฐ์— ์˜ํ•ด ์ถ”์ •๋œ ํ‘œ์  ์…€(cell)์— ๋Œ€ํ•œ ํƒ์ง€ ๊ฒฐ๊ณผ๋ฅผ ์ด์šฉํ•˜๋ฉด ์‹ (18)๊ณผ ๊ฐ™์ด SCR๋ฅผ ์ถ”์ •ํ•  ์ˆ˜ ์žˆ๋‹ค.

(18)
$\hat SCR =\left | P_{p}-(Th_{CFAR}-\alpha_{CFAR})\right |_{t\arg et cell}[d B]$

์—ฌ๊ธฐ์„œ, $P_{p}$๋Š” ํ‘œ์  ์…€์˜ ์‹ ํ˜ธ ์ „๋ ฅ, $Th_{CFAR}$๋Š” ํ‘œ์  ์…€์— ๋Œ€ํ•œ CFAR ํƒ์ง€ ์ž„๊ณ„๊ฐ’, $\alpha_{CFAR}$๋Š” CFAR ํƒ์ง€ ์ž„๊ณ„๊ฐ’์„ ์œ„ํ•œ ์ƒ์ˆ˜๋ฅผ ์˜๋ฏธํ•œ๋‹ค.

๋ ˆ์ด๋” ์‹œ์Šคํ…œ๊ณผ ํ‘œ์ ๊ฐ„์˜ ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ๋Š” ์‹œ๊ฐ„์ด ์ง€๋‚จ์— ๋”ฐ๋ผ ์ƒ๋Œ€ ์†๋„์— ์˜ํ•˜์—ฌ ์ค„์–ด๋“ ๋‹ค๋Š” ์ „์ œ ์กฐ๊ฑดํ•˜์—, ์ดˆ๊ธฐ ํƒ์ƒ‰ ๋‹จ๊ณ„์—์„œ ์‹ (18)์— ์˜ํ•œ SCR ์ถ”์ •๊ฐ’์ด SCR ์˜ˆ์ƒ๊ฐ’๋ณด๋‹ค ์ž‘๋‹ค๋ฉด $P_{D,\:P}$๊ฐ€ $P_{D,\:D}$๋ณด๋‹ค ์ž‘๋‹ค๊ณ  ํŒ๋‹จํ•  ์ˆ˜ ์žˆ๋‹ค. ๊ทธ๋ฆฌ๊ณ  ํƒ์ƒ‰ ์‹œ๊ฐ„์ด ์ง€๋‚จ์— ๋”ฐ๋ผ SCR์ด ์ฆ๊ฐ€ํ•˜๊ฒŒ ๋˜๋ฉด $P_{D,\:P}$๊ฐ€ $P_{D,\:D}$์™€ ๊ฐ™์•„์ง€๊ฒŒ ๋  ๊ฒƒ์ด๋ฏ€๋กœ SCR ์ถ”์ •๊ฐ’๊ณผ SCR ์˜ˆ์ƒ๊ฐ’์˜ ์ฐจ์ด์— ํ•ด๋‹นํ•˜๋Š” ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ ์ฐจ์ด๋Š” ์‹ (17)์— ์˜ํ•ด ์‚ฐ์ถœ๋  ์ˆ˜ ์žˆ๋‹ค. ๋”ฐ๋ผ์„œ ํ‘œ์ ์ด ์กด์žฌํ•˜์ง€๋งŒ ํ‘œ์  ์ธก์ •์น˜๊ฐ€ ๊ด€์ฐฐ๋˜์ง€ ์•Š๋Š” ๊ตฌ๊ฐ„์„ ๊ฒฐ์ •ํ•˜๋Š” $m$ ์‹œ์ ์— ๋Œ€ํ•œ ๊ธฐ๋Œ“๊ฐ’์€ ์‹ (19)์™€ ๊ฐ™์ด ํ‘œํ˜„๋  ์ˆ˜ ์žˆ๋‹ค.

(19)
$E[m |\chi^{n}]=roun d\left(\dfrac{\Delta R_{r}}{V_{c}\Delta t}\right)$

์—ฌ๊ธฐ์„œ, $\triangle R_{r}$๋Š” SCR ์ฐจ์ด์— ์˜ํ•ด ์‚ฐ์ถœ๋˜๋Š” ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ ์ฐจ์ด, $V_{c}$๋Š” ๋ ˆ์ด๋” ์‹œ์Šคํ…œ๊ณผ ํ‘œ์ ๊ฐ„์˜ ์ƒ๋Œ€ ์†๋ ฅ, $\triangle t$๋Š” ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ์‹ ํ˜ธ ์ฒ˜๋ฆฌ ์ฃผ๊ธฐ๋ฅผ ์˜๋ฏธํ•œ๋‹ค.

๊ทธ๋Ÿฐ๋ฐ, ์ดˆ๊ธฐ ํƒ์ƒ‰ ๋‹จ๊ณ„์—์„œ SCR์ด ๋งค์šฐ ๋‚ฎ์•„ ํ‘œ์  ์‹ ํ˜ธ๊ฐ€ ํƒ์ง€๋˜์ง€ ์•Š๋Š” ๊ฒฝ์šฐ์—๋Š” ์ถ”์  ํ•„ํ„ฐ๊ฐ€ ์ •์ƒ์ ์ธ ์ดˆ๊ธฐํ™” ๊ณผ์ •์„ ์ˆ˜ํ–‰ํ•˜์ง€ ๋ชปํ•˜์—ฌ ํ‘œ์  ์…€์„ ์ถ”์ •ํ•  ์ˆ˜ ์—†์„ ๊ฒƒ์ด๋‹ค. ์ด๋Ÿฌํ•œ ์ƒํ™ฉ์—์„œ๋Š” SCR์ด ์ฆ๊ฐ€๋˜๋„๋ก ํ•ด๋‹น ์˜์—ญ์— ๋Œ€ํ•œ ํƒ์ƒ‰์„ ์ง€์†ํ•˜๊ฑฐ๋‚˜ ํ•ด๋‹น ์˜์—ญ์— ํ‘œ์ ์ด ์กด์žฌํ•˜์ง€ ์•Š๋Š”๋‹ค๊ณ  ๊ฐ„์ฃผํ•˜์—ฌ ๋‹ค๋ฅธ ์˜์—ญ์„ ํƒ์ƒ‰ํ•˜๋Š” ์šด์šฉ ๊ฐœ๋…์„ ์‹œ์Šคํ…œ ์„ค๊ณ„์— ๋ฐ˜์˜ํ•˜๋ฉด ๋œ๋‹ค.

4. ์‹œ๋ฎฌ๋ ˆ์ด์…˜ ๋ฐ ๊ฒฐ๊ณผ๋ถ„์„

๋ณธ ์žฅ์—์„œ๋Š” ์ด๋™ํ˜• ํ”Œ๋žซํผ์— ํƒ‘์žฌ๋œ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์ด ์ง€์ƒ ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์˜ ๋‹จ์ผ ํ‘œ์ ์„ ํƒ์ƒ‰ํ•˜๋Š” ์šด์šฉ ์กฐ๊ฑด์—์„œ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์„ ์˜ˆ์ธกํ•˜๋Š” ๊ธฐ๋ฒ•์— ๋Œ€ํ•œ ์‹œ๋ฎฌ๋ ˆ์ด์…˜ ๊ฒฐ๊ณผ๋ฅผ ๋ณด์—ฌ์ค€๋‹ค. ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ์˜ ๊ฐœ์ˆ˜๋Š” ํด๋Ÿฌํ„ฐ ์ธก์ •์น˜ ๋ฐ€๋„(Clutter Measurement Density ; CMD) ๊ธฐ์ค€ ํฌ์•„์†ก(Poisson) ๋ถ„ํฌ์— ์˜ํ•ด ์‹œ๊ฐ„์— ๋”ฐ๋ผ ๋ณ€๊ฒฝ๋˜๊ณ , ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ์˜ ์œ„์น˜๋Š” ํƒ์ƒ‰ ์˜์—ญ์—์„œ ๊ท ์ผ ๋ถ„ํฌ์— ์˜ํ•ด ๋žœ๋ค(random)ํ•˜๊ฒŒ ์ƒ์„ฑ๋˜๋Š” ํ™˜๊ฒฝ์— ๋Œ€ํ•˜์—ฌ, ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์€ ๊ธฐ๋ณธ์ ์œผ๋กœ ์ง€์ •๋œ ํƒ์ƒ‰ ์˜์—ญ์—์„œ ํ‘œ์  ํƒ์ƒ‰์„ ์‹œ์ž‘ํ•œ๋‹ค. ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ๊ฐ•์ธํ•œ ํ‘œ์  ํƒ์ง€ ์„ฑ๋Šฅ์„ ๋ณด์žฅํ•ด ์ค„ ์ˆ˜ ์žˆ๋Š” OS(Order Statistics) CFAR ํƒ์ง€ ๊ธฐ๋ฒ•์„ ์ ์šฉํ•˜์—ฌ ์ˆ˜์‹  ์‹ ํ˜ธ๋ฅผ ํƒ์ง€ํ•œ๋‹ค. ๊ทธ๋ฆฌ๊ณ  ํš๋“๋˜๋Š” ๋‹ค์ˆ˜ ์ธก์ •์น˜์— ๋Œ€ํ•˜์—ฌ PTE ๊ธฐ๋ฐ˜ ์ถ”์  ํ•„ํ„ฐ์ธ IPDA(Integrated Probabilistic Data Association) ํ•„ํ„ฐ๋ฅผ ์ด์šฉํ•˜์—ฌ ํ‘œ์ ์˜ ์œ„์น˜๋ฅผ ์ถ”์ ํ•˜๋ฉด์„œ PTE๋ฅผ ๊ฐฑ์‹ ํ•œ๋‹ค. ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ์œ„ํ•œ ๊ณตํ†ต์ ์ธ ์„ค๊ณ„ ๋ณ€์ˆ˜๋Š” ํ‘œ 1๊ณผ ๊ฐ™๊ณ , ๊ฐ ์‹œ๋ฎฌ๋ ˆ์ด์…˜ ์กฐ๊ฑด์— ๋Œ€ํ•œ ํ†ต๊ณ„์ ์ธ ๋ถ„์„์„ ์œ„ํ•ด 100ํšŒ์”ฉ ๋ชฌํ…Œ์นด๋ฅผ๋กœ(Monte-Carlo) ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ์ˆ˜ํ–‰ํ•˜์˜€๋‹ค.

ํ‘œ 1 ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ์œ„ํ•œ ๊ณตํ†ต ์„ค๊ณ„ ๋ณ€์ˆ˜

Table 1 Common design parameters for simulation

Parameter

Value

Parameter

Value

$P_{D}$

0.9

$\Pi$

$$ \left[\begin{array}{lll} 0.90 & 0.05 & 0.05 \\ 0.05 & 0.90 & 0.05 \\ 0.00 & 0.00 & 1.00 \end{array}\right] $$

$P_{G}$

0.99

$T_{C}$

0.9

$\triangle t$

0.1 sec.

๊ทธ๋ฆผ 2๋Š” ํ‘œ์  ์‹ ํ˜ธ์— ๋Œ€ํ•œ ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ์˜ ์˜ํ–ฅ์— ๋”ฐ๋ฅธ PTE ์„ฑ๋Šฅ ๋น„๊ต ๊ฒฐ๊ณผ๋ฅผ ๋ณด์—ฌ์ค€๋‹ค. ํ‘œ์ ์ด ์กด์žฌํ•˜๋Š” ์˜์—ญ์˜ CMD๊ฐ€ ๋‚ฎ์€ ๊ฒฝ์šฐ์—๋Š” ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ ์„ฑ๋ถ„์ด ํ‘œ์  ์‹ ํ˜ธ๊ฐ€ ์กด์žฌํ•˜๋Š” ์…€์— ์˜ํ–ฅ์„ ์ฃผ์ง€ ์•Š์•„ PTE๊ฐ€ ์›ํ•˜๋Š” ์ž„๊ณ„๊ฐ’($T_{C}$)์— ๋„๋‹ฌํ•˜๋Š” ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์ด ์‹ (6)์— ์˜ํ•ด ์‚ฐ์ถœ๋˜๋Š” ์˜ˆ์ธก๊ฐ’(0.2์ดˆ)์„ ๋งŒ์กฑํ•˜์ง€๋งŒ, CMD๊ฐ€ ๋†’์€ ๊ฒฝ์šฐ์—๋Š” ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ ์„ฑ๋ถ„์ด ํ‘œ์  ์‹ ํ˜ธ๊ฐ€ ์กด์žฌํ•˜๋Š” ์…€์— ์ง์ ‘์  ๋˜๋Š” ๊ฐ„์ ‘์ ์œผ๋กœ ์˜ํ–ฅ์„ ์ฃผ๊ฒŒ ๋˜์–ด ์‹ค์ œ SCR์ด ๋‚ฎ์•„์ง€๊ฒŒ ๋˜๋ฉฐ, ์ด์— ๋”ฐ๋ผ PTE์— ์˜ํ•œ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์ด ์‹ (6)์— ์˜ํ•œ ์˜ˆ์ธก๊ฐ’๋ณด๋‹ค ๋งŽ์ด ์†Œ์š”๋˜๋Š” ๊ฒƒ(์•ฝ 1.5์ดˆ)์„ ๋ณผ ์ˆ˜ ์žˆ๋‹ค.

๊ทธ๋ฆผ 2 CMD์— ๋”ฐ๋ฅธ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์„ฑ๋Šฅ; (a) CMD= 2e-5์ผ ๋•Œ ํƒ์ง€๊ฒฐ๊ณผ, (b) CMD=2e-3์ผ ๋•Œ ํƒ์ง€๊ฒฐ๊ณผ, (c) CMD์— ๋”ฐ๋ฅธ PTE ๊ฒฐ๊ณผ

Fig. 2 Performance of target lock-on time for each CMD; (a) Detection result at CMD=2e-5, (b) Detection result at CMD=2e-3, (c) PTE results for each CMD

../../Resources/kiee/KIEE.2022.71.4.663/fig2.png

๊ทธ๋ฆผ 3์€ ๋‹ค์–‘ํ•œ ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ์ œ์•ˆํ•˜๋Š” ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์˜ˆ์ธก ๊ธฐ๋ฒ•์˜ ์„ฑ๋Šฅ์„ ๋ณด์—ฌ์ค€๋‹ค. ์ด๋™ํ˜• ํ”Œ๋žซํผ์— ํƒ‘์žฌ๋œ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์„ ์˜ˆ์ธกํ•˜๊ธฐ ์œ„ํ•˜์—ฌ ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์ด ํƒ์ƒ‰ ์˜์—ญ์˜ ์ค‘์‹ฌ ๊ฑฐ๋ฆฌ์ธ 5 km๋ฅผ ๊ธฐ์ค€์œผ๋กœ ๋น” ์ œํ•œ์  ์šด์šฉ ์กฐ๊ฑด์—์„œ 500 m/s ์ƒ๋Œ€ ์†๋„๋กœ ํ‘œ์ ๊ณผ์˜ ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ๊ฐ€ ์ค„์–ด๋“ ๋‹ค๊ณ  ๊ฐ€์ •ํ•˜์˜€๊ณ , ์ด์— ๋”ฐ๋ผ SCR์€ 1์ดˆ๋งˆ๋‹ค 1 dB์”ฉ ์ฆ๊ฐ€๋œ๋‹ค๊ณ  ๊ทผ์‚ฌํ™”ํ•˜์˜€๋‹ค. ํ‘œ 2์™€ ๊ฐ™์ด ํด๋Ÿฌํ„ฐ ์˜ํ–ฅ์ด ์—†๋Š” ์ด์ƒ์ ์ธ ์กฐ๊ฑด์ธ CMD = 2e-5์—์„œ ์ถ”์ •ํ•œ ํ‘œ์  SCR๋ฅผ ๊ธฐ์ค€์œผ๋กœ ๋‹ค์–‘ํ•œ CMD ์กฐ๊ฑด์—์„œ ์‹ (18)์— ์˜ํ•ด ์ถ”์ •ํ•œ ํ‘œ์  SCR๊ณผ์˜ ์ฐจ์ด($\triangle SCR$)๋ฅผ ์‚ฐ์ถœํ•˜๊ณ , ์‹ (17)์— ์˜ํ•ด $\triangle SCR$์ด ์ƒ์‡„๋˜๊ธฐ ์œ„ํ•œ ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ ์ฐจ์ด($\triangle R_{r}$)๋ฅผ ๊ณ„์‚ฐํ•˜๋ฉฐ, ์ด๋ฅผ ๊ธฐ์ค€์œผ๋กœ ์‹ (16)์— ์˜ํ•œ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์„ ์˜ˆ์ธกํ•˜์˜€๋‹ค. CMD๊ฐ€ ๋†’์•„์งˆ์ˆ˜๋ก ์ƒ๋Œ€ ๊ฑฐ๋ฆฌ ์ฐจ์ด๊ฐ€ ์ปค์ง€๊ฒŒ ๋˜์–ด ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์ด ์ƒ๋Œ€์ ์œผ๋กœ ๊ธธ์–ด์ง€๋Š” ํŠน์„ฑ์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋Š”๋ฐ, ์ด๋Š” CMD๊ฐ€ ๋†’์•„์งˆ์ˆ˜๋ก ํ‘œ์  ์‹ ํ˜ธ์— ๋Œ€ํ•œ ํด๋Ÿฌํ„ฐ ์‹ ํ˜ธ ์„ฑ๋ถ„์˜ ์˜ํ–ฅ์ด ์ปค์ง€๊ฒŒ ๋˜์–ด ๊ฒฐ๊ณผ์ ์œผ๋กœ SCR์ด ๋‚ฎ์•„์กŒ๊ณ , ์ด์— ๋”ฐ๋ผ SCR์ด ์ฆ๊ฐ€ํ•˜๋Š” ์‹œ๊ฐ„์ด ๊ธธ์–ด์ง„ ๊ฒƒ์œผ๋กœ ๋ถ„์„ํ•  ์ˆ˜ ์žˆ๋‹ค. ๊ทธ๋ฆฌ๊ณ  ํ‘œ 2์™€ ๊ฐ™์ด ๋‹ค์–‘ํ•œ CMD์— ๋Œ€ํ•˜์—ฌ ์‹œ๋ฎฌ๋ ˆ์ด์…˜์— ์˜ํ•œ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์ด ์ด๋ก ์ ์œผ๋กœ ์˜ˆ์ธกํ•œ ํฌ์ฐฉ ์‹œ๊ฐ„๊ณผ ๊ฑฐ์˜ ์œ ์‚ฌํ•จ์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค.

๊ทธ๋ฆผ 3 CMD์— ๋”ฐ๋ฅธ ์ œ์•ˆ ๊ธฐ๋ฒ•์˜ PTE ๊ฒฐ๊ณผ

Fig. 3 PTE result of proposed method for each CMD

../../Resources/kiee/KIEE.2022.71.4.663/fig3.png

ํ‘œ 2 ์ œ์•ˆ ๊ธฐ๋ฒ•์˜ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์˜ˆ์ธก ๊ฒฐ๊ณผ

Table 2 Results of proposed method for predicting target lock-on time

CMD

$\triangle SCR$

[dB]

$\triangle R_{r}$

[m]

Expected Time [sec.]

Simulated Time [sec.]

2e-5

0.0

0

0.2

0.2

1e-3

-0.5

250

0.7

0.5

2e-3

-1.0

500

1.2

1.2

3e-3

-2.0

1,000

2.2

2.3

5e-3

-4.0

2,000

4.2

4.1

5. ๊ฒฐ ๋ก 

ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ์šด์šฉํ•˜๋Š” ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์˜ ์„ฑ๋Šฅ์€ ๊ธฐ๋ณธ์ ์œผ๋กœ ํด๋Ÿฌํ„ฐ์— ์˜ํ–ฅ์„ ๋ฐ›๊ฒŒ ๋˜๊ณ , ํšจ์œจ์ ์ธ ํ‘œ์  ํƒ์ƒ‰์„ ์ˆ˜ํ–‰ํ•˜๊ธฐ ์œ„ํ•ด์„œ๋Š” ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์— ๋Œ€ํ•œ ์˜ˆ์ธก ๊ธฐ๋ฒ•์ด ํ•„์š”ํ•˜๋‹ค. ๊ธฐ์กด์˜ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„ ์˜ˆ์ธก ๊ธฐ๋ฒ•์€ ํ‘œ์  ์‹ ํ˜ธ์˜ ํƒ์ง€ ํ™•๋ฅ ์„ ์‚ฌ์ „์— ์•Œ๊ณ  ์žˆ์–ด์•ผ ๋˜๋Š”๋ฐ, ํด๋Ÿฌํ„ฐ ํŠน์„ฑ์€ ์‹œ๊ฐ„๊ณผ ๊ณต๊ฐ„์— ๋”ฐ๋ผ ํฌ๊ฒŒ ๋ณ€๊ฒฝ๋  ์ˆ˜ ์žˆ๊ธฐ์— ํƒ์ง€ ํ™•๋ฅ ์„ ์ •ํ™•ํžˆ ์•Œ ์ˆ˜ ์—†๋Š” ํ•œ๊ณ„๊ฐ€ ์žˆ๋‹ค. ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ํš๋“๋˜๋Š” ํƒ์ง€ ๊ฒฐ๊ณผ๋ฅผ ์ด์šฉํ•œ ํ‘œ์  SCR ์ถ”์ • ๋ฐฉ๋ฒ•๊ณผ ๋งˆ์ฝ”ํ”„ ์ฒด์ธ-2 ๋ชจ๋ธ์„ ์ด์šฉํ•œ PTE ๋ฐฉ๋ฒ•์„ ๊ธฐ๋ฐ˜์œผ๋กœ ํ‘œ์  ํฌ์ฐฉ ์‹œ๊ฐ„์„ ์˜ˆ์ธกํ•˜๋Š” ๊ธฐ๋ฒ•์„ ์ œ์•ˆํ•˜๊ณ , ์ปดํ“จํ„ฐ ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ํ†ตํ•ด ํด๋Ÿฌํ„ฐ ํ™˜๊ฒฝ์—์„œ ์šด์šฉ๋˜๋Š” ๋ ˆ์ด๋” ์‹œ์Šคํ…œ์— ๋Œ€ํ•œ ์ œ์•ˆ ๊ธฐ๋ฒ•์˜ ์ ์šฉ ๊ฐ€๋Šฅ์„ฑ์„ ํ™•์ธํ•˜์˜€๋‹ค.

Acknowledgements

References

1 
M. I. Skolnik, 2008, Radar Handbook, McGraw-Hill EducationGoogle Search
2 
M. A. Richards, J. A. Scheer, W. A. Holm, 2012, Principles of Modern Radar, Scitech PublishingGoogle Search
3 
C. Alabaster, 2012, Pulse Doppler Radar, Scitech PublishingGoogle Search
4 
D. Musicki, R. Evans, S. Stankovic, 1994, Integrated Probabilistic Data Association (IPDA), IEEE Trans. Autom. Control, Vol. 39, No. 6, pp. 1237-1241DOI
5 
X. Wang, D. Musicki, 2003, Low Elevation Sea-Surface Target Tracking using IPDA Type Filters, 2003 Proceedings of the International Conference on Radar(IEEE Cat. no. 03EX695). IEEE, pp. 472-478DOI
6 
S. Challa, M. R. Morelande, D. Musicki, R. J. Evans, 2011, Fundamental of Object Tracking, Cambridge University PressGoogle Search
7 
D. Musicki, 2006, Track Score and Target Existence, 2006 9th International Conference on Information Fusion. IEEE, pp. 1-7DOI
8 
T. L. Song, D. Musicki, 2010, Target Existence Based Resource Allocation, IEEE Trans. Signal Process., Vol. 58, No. 9, pp. 4496-4506DOI
9 
T. L. Lane, C. R. Scheer, N. C. Currie, 1988, An Overview of MMW Radar Reflectivity, Proceedings of the 1988 IEEE National Radar Conference. IEEE, pp. 235-240DOI
10 
N. C. Currie, S. W. Parker, R. B. Efurd, 1988, MMW System Trade-offs, IEEE Trans. Aerosp. Electron. Syst. Mag., Vol. 3, No. 10, pp. 21-26DOI
11 
F. E. Nathanson, J. P. Reilly, M. N. Cohen, 1999, Radar Design Principles, Scitech PublishingGoogle Search
12 
M. W. Long, 2001, Radar Reflectivity of Land and Sea (3rd), Artech HouseGoogle Search
13 
D. Musicki, R. Evans, 2004, Clutter Map Information for Data Association and Track Initialization, IEEE Trans. Aerosp. Electron. Syst., Vol. 40, No. 2, pp. 387-398DOI
14 
Y. Kim, T. L. Song, 2013, A Study of Image Target Tracking Using ITS in an Occluding Environment, Journal of Institute of Control, Vol. robotics and systems, pp. 306-314DOI
15 
T. L. Song, D. Musicki, Y. Kim, 2013, Tracking through Occlusions and Track Segmentation Reduction, IEEE Trans. Aerosp. Electron. Syst., Vol. 49, No. 1, pp. 623-631DOI

์ €์ž์†Œ๊ฐœ

์‹ ์ •ํ›ˆ (Jeong-Hoon Shin)
../../Resources/kiee/KIEE.2022.71.4.663/au1.png

1999๋…„ ์ถฉ๋‚จ๋Œ€ํ•™๊ต ์ „์ž๊ณตํ•™๊ณผ ์กธ์—…(ํ•™์‚ฌ)

2001๋…„ ์ถฉ๋‚จ๋Œ€ํ•™๊ต ์ „์ž๊ณตํ•™๊ณผ ์กธ์—…(์„์‚ฌ)

2022๋…„ ํ•œ์–‘๋Œ€ํ•™๊ต ์ „์ž์‹œ์Šคํ…œ๊ณตํ•™๊ณผ ์กธ์—…(๋ฐ•์‚ฌ)

2002๋…„โˆผํ˜„์žฌ ๊ตญ๋ฐฉ๊ณผํ•™์—ฐ๊ตฌ์†Œ

๊ด€์‹ฌ๋ถ„์•ผ : ๋ ˆ์ด๋” ์‹ ํ˜ธ์ฒ˜๋ฆฌ, ํ‘œ์  ์ถ”์ ํ•„ํ„ฐ

์ตœ์˜์ง„ (Youngjin Choi)
../../Resources/kiee/KIEE.2022.71.4.663/au2.png

1994๋…„ ํ•œ์–‘๋Œ€ํ•™๊ต ์ •๋ฐ€๊ธฐ๊ณ„๊ณตํ•™๊ณผ ์กธ์—…(ํ•™์‚ฌ)

1996๋…„ ํฌํ•ญ๊ณต๊ณผ๋Œ€ํ•™๊ต ๊ธฐ๊ณ„๊ณตํ•™๊ณผ ์กธ์—…(์„์‚ฌ)

2002๋…„ ํฌํ•ญ๊ณต๊ณผ๋Œ€ํ•™๊ต ๊ธฐ๊ณ„๊ณตํ•™๊ณผ ์กธ์—…(๋ฐ•์‚ฌ)

2005๋…„โˆผํ˜„์žฌ ํ•œ์–‘๋Œ€ํ•™๊ต ์ „์ž๊ณตํ•™๋ถ€ ๊ต์ˆ˜

๊ด€์‹ฌ๋ถ„์•ผ : ๋กœ๋ด‡, ์ œ์–ด, ์ƒ์ฒด์‹ ํ˜ธ์ฒ˜๋ฆฌ

์†กํƒ๋ ฌ (Taek-Lyul Song)
../../Resources/kiee/KIEE.2022.71.4.663/au3.png

1974๋…„ ์„œ์šธ๋Œ€ํ•™๊ต ๊ณตํ•™์‚ฌ

1981๋…„ University of Texas at Austin ๋Œ€ํ•™์› ํ•ญ๊ณต์šฐ์ฃผ๊ณตํ•™๊ณผ ์กธ์—…(์„์‚ฌ)

1983๋…„ University of Texas at Austin ๋Œ€ํ•™์› ํ•ญ๊ณต์šฐ์ฃผ๊ณตํ•™๊ณผ ์กธ์—…(๋ฐ•์‚ฌ)

1974๋…„โˆผ1995๋…„ ๊ตญ๋ฐฉ๊ณผํ•™์—ฐ๊ตฌ์†Œ

1995๋…„~2017๋…„(8์›”) ํ•œ์–‘๋Œ€ํ•™๊ต ์ „์ž๊ณตํ•™๋ถ€ ๊ต์ˆ˜

2017๋…„(9์›”)~ํ˜„์žฌ ํ•œ์–‘๋Œ€ํ•™๊ต ์ „์ž๊ณตํ•™๋ถ€ ๋ช…์˜ˆ๊ต์ˆ˜

๊ด€์‹ฌ๋ถ„์•ผ : ํ‘œ์ ์ถ”์  ์‹œ์Šคํ…œ, ์ž๋ฃŒ์—ฐ๊ด€, ์ •๋ณด์œตํ•ฉ, ์œ ๋„ ๋ฐ ์ œ์–ด