-
2023.09 Sunday Monday Tuesday Wednesday Thursday Friday Saturday 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Manager
조회 28203
Status : | Presented |
---|---|
Date : | 1991-11 |
Title : | TFM 방식에서 Trellis 특성을 이용한 Nonredundant 에러 정정 |
Authors : | 홍대식, 강창언, 조형래, 정의성 |
Conference : | 한국통신학회 추계학술발표회 논문집 10권 3호 |
Abstract : | In this paper, trellis detection scheme is proposed to improve the performance of the noncoherent detection in the TFM system. Trellis detection takes advantage of the trellis property of TFM-encoded signal. The trellis property is created by giving correlations among adjacent TFM-encoded signals at the transmitter. To apply trellis detection, noncoherent frequency discriminator demodulation and 5-level hard-decision are assumed. At the receiver, the received signal outputs from 5-level hard-decision are applied to the detection algorithm, and NRZ data are detected. The detection algorithm in the trellis detection is named trellis detection algorithm (TDA). TDA is a kind of maximum likelihood sequence estimation (MLSE). Performance of the trellis detection scheme is analyzed by means of Bernoulli trails with average symbol error probability, and is compared to that of bit-by-bit detection scheme. As a result, when SNR is below 20 dB, trellis detection is inferior to the bit-by-bit detection. But when SNR is above 20 dB, trellis detection is superior to the bit-by-bit detection, and its performance enhancement is better as the SNR increased. |
URL : | http://mirinae.yonsei.ac.kr//?module=fil...f3bf27cabd |
-
2023.09 Sunday Monday Tuesday Wednesday Thursday Friday Saturday 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
-
2023.09 Sunday Monday Tuesday Wednesday Thursday Friday Saturday 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30