Page 72 - 《应用声学》2022年第1期
P. 72
68 2022 年 1 月
0 并且在复杂水声信道下,所提方法能够取得比在
20 AWGN 信道下更高的信噪比增益。湖上实验结果
-5
15 显示,所提方法能够将系统误码率由传统分离译码
-10 的 1.9×10 −2 降低为 4.0×10 −4 。当传输图像序列时,
ᫎ/s 10 -15 ॆʷӑη᥋־ऄ/dB 所提方法对全部接收信号均实现了无误码通信,而
5 传统分离译码的误码率仍然为3.7×10 −3 。
-20
在后续的研究当中,一方面,考虑继续细化分
-25 析译码参数 α 对译码性能的影响,研究效率和精度
-20 0 20 40 60
ኀՂᫎᬦ 更高的参数选取方法,进一步提高水声通信系统的
(a) ԫη᥋фѤ־ऄͥᝠፇ౧
可靠性;另一方面,进一步优化信源先验信息的提取
0.6
方法,提高信源符号间相关性的利用效率。
0.5
0.4
ࣨए 0.3 参 考 文 献
0.2
[1] Steele R, Goodman D J. Detection and selective smooth-
0.1
ing of transmission errors in linear PCM[J]. Bell System
0 Technical Journal, 1977, 56(3): 399–409.
-20-10 0 10 20 30 40 50 60 70 80
ኀՂᫎᬦ [2] Sayood K, Borkenhagen J C. Use of residual redundancy
in the design of joint source/channel coders[J]. IEEE
(b) ᅯη᥋фѤ־ऄͥᝠፇ౧
Transactions on Communications, 1991, 39(6): 838–846.
图 10 实测信道冲击响应 [3] Demir N, Sayood K. Joint source/channel coding for vari-
Fig. 10 Measured channel impulse response able length codes[C]//Proceedings DCC’98 Data Com-
pression Conference. Snowbird, USA, 1998: 139–148.
[4] Park M, Miller D J. Joint source-channel decoding for
variable-length encoded data by exact and approximate
MAP sequence estimation[J]. IEEE Transactions on Com-
munications, 2000, 48(1): 1–6.
[5] Murad A H, Fuja T E. Joint source-channel decoding
of variable-length encoded sources[C]//1998 Information
Theory Workshop. Killarney, Ireland, 1998: 94–95.
[6] Lakovic K, Villasenor J, Wesel R. Robust joint Huffman
and convolutional decoding[C]//Gateway to 21st Century
Communications Village. VTC 1999-Fall. IEEE VTS
(a) ͜ፒѬሏឋᆊፇ౧ (b) ਫ਼ଢவขឋᆊፇ౧
50th Vehicular Technology Conference. Amsterdam, The
图 11 译码结果 Netherlands, 1999, 5: 2551–2555.
Fig. 11 Decoding results [7] Bauer R, Hagenauer J. Iterative source/channel-decoding
using reversible variable length codes[C]//Proceedings
DCC 2000. Data Compression Conference. IEEE, Snow-
5 结论
bird, UT, USA, 2000: 93–102.
[8] Berrou C, Glavieux A, Thitimajshima P. Near Shan-
本文根据 Polar 码的译码结构提出了一种基于 non limit error-correcting coding and decoding: Turbo-
Polar 码的水声通信信源信道联合译码方法。该方 codes. 1[C]//Proceedings of ICC ’93-IEEE International
法以信源状态转移关系为基础构建信源信道联合 Conference on Communications, Geneva, Switzerland,
1993, 2: 1064–1070.
译码网格图,在联合译码网格图上同时进行信源译
[9] Gallager R. Low-density parity-check codes[J]. IRE Trans-
码和信道译码,综合信源转移概率和信道转移概率 actions on Information Theory, 1962, 8(1): 21–28.
计算统一的后验概率,实现了信源译码和信道译码 [10] Jeanne M, Carlach J C, Siohan P, et al. Source and joint
的一体化联合优化。仿真结果表明,所提方法能够 source-channel decoding of variable length codes[C]//2002
IEEE International Conference on Communications. Con-
利用信源残留冗余抵抗信道差错,在不降低通信速
ference Proceedings. ICC 2002. New York, USA, 2002, 2:
率的情况下,取得优于传统分离译码的误码率性能, 768–772.