Systems Engineering and Electronics ›› 2026, Vol. 48 ›› Issue (5): 1752-1764.doi: 10.12305/j.issn.1001-506X.2026.05.30

• Communications and Networks • Previous Articles     Next Articles

High-order modulation and soft detection of multi-carrier underwater acoustic communication with colored noise

Rui XU1,2,3(), Yanbo WU1,3,4,*(), Min ZHU1,3,4()   

  1. 1. Ocean Acoustic Technology Center,Institute of Acoustics,Chinese Academy of Sciences,Beijing 100190,China
    2. School of Electronic,Electrical and Communication Engineering,University of Chinese Academy of Sciences,Beijing 100049,China
    3. Beijing Engineering Technology Research Center of Ocean Acoustic Equipment,Beijing 100190,China
    4. State Key Laboratory of Acoustics,Institute of Acoustics,Chinese Academy of Sciences,Beijing 100190,China
  • Received:2024-12-30 Online:2026-05-27 Published:2026-05-27
  • Contact: Yanbo WU E-mail:xurui182@mail.ucas.edu.cn;wuyanbo@ioa.ac.cn;zhumin@mail.ioa.ac.com

Abstract:

In the field of underwater acoustic communication, orthogonal frequency division multiplexing (OFDM) combined with bit-interleaved coded modulation technology can effectively mitigate frequency-selective fading in underwater acoustic channels. However, due to incomplete acquisition of channel state information and noise interference, higher-order modulation schemes aimed at achieving higher communication rates are more sensitive to channel mismatches. To enhance decoding performance and communication reliability, this study delves into the statistical characteristics of channel estimation errors and proposes a maximum likelihood-based decoding metric design, which modifies the likelihood function used in the decoding process. Simultaneously, considering the characteristics of real-world noise, the channel estimation error correction decoding metric is further optimized under the colored noise assumption. The method estimates noise information for each subcarrier with the colored noise hypothesis and corrects the channel estimates based on the statistical characteristics of channel estimation errors, thereby reducing the impact of channel estimation errors on the output soft information in the presence of colored noise. The simulation results show that the color noise assumption (channel estimation error correction) decoding metric achieves a higher performance upper bound and offers a 5-6 dB improvement over the traditional (channel estimation error ignored) decoding metric, while providing a 1.4 dB enhancement compared to the white noise assumption (channel estimation error correction) decoding metric. Under the conditions of the deepest 900 m deep-sea vertical communication test, it was verified that the noise spectral density fluctuation in the frequency domain can reach 12 dB. The proposed method effectively improves decoding performance for OFDM signals with up to 256-order quadrature amplitude modulation (QAM) and a transmission rate of 8.84 kbps. At a communication distance of up to 4772 m, reliable OFDM underwater acoustic communication was achieved with a transmission rate of 4.42 kbps using 16QAM modulation.

Key words: underwater acoustic communication, orthogonal frequency division multiplexing (OFDM), colored noise, channel estimation, log-likelihood ratio

CLC Number: 

[an error occurred while processing this directive]