Elsevier

Applied Acoustics

Volume 173, February 2021, 107654
Applied Acoustics

Subcarrier modulation identification of underwater acoustic OFDM based on block expectation maximization and likelihood

https://doi.org/10.1016/j.apacoust.2020.107654Get rights and content

Highlights

  • A subcarrier modulation identification method based on Expectation Maximization and Likelihood is proposed. The method of Expectation Maximization is used in channel estimation and recovery constellation and the method of Likelihood is used in identification.

  • According to the structural characteristics of underwater acoustic OFDM, a blind channel estimation method based on Block Expectation Maximization is proposed, it improves the performance of channel estimation and greatly improves the identification rate.

Abstract

The low identification rate of the Orthogonal frequency division multiplexing (OFDM) based subcarrier modulation in underwater acoustic multipath channel is an important issue. Therefore, we proposed a novel Expectation Maximization-Block-Quasi Hybrid Likelihood Ratio Test (EM-Block-QHLRT) method which effectively improved the identification rate while using the blind channel impulse response (CIR) estimation and likelihood. Initially, CIR is obtained by using clustering, and then the CIR is updated iteratively by EM-Block method. Further, the subcarrier modulation is identified using QHLRT. The influence of iteration times, the length of symbol and influence of the number of blocks in EM are analyzed by using extensive simulation. The identification rate of subcarrier modulation of binary phase shift keying (BPSK), QPSK, 8PSK and 16QAM are presented using different signal noise ratio (SNR). In addition, the identification rate of subcarrier modulation based on Average Likelihood Rate Test (ALRT) as well as QHLRT with EM are compared. Simulation results showed that the identification rate of the proposed EM-Block-QHLRT method can reach to more than 90% when SNR is higher than 5 dB. Finally, the performance of the proposed EM-Block-QHLRT method is verified using sea trial data.

Introduction

OFDM has been widely implemented in underwater acoustic communication systems because of its high spectral efficiency and good performance in combating interference, and has become the prominent technology for high-speed underwater acoustic communication systems [1], [2], [3], [4]. Further, research on Non-Cooperative underwater acoustic OFDM has become a hotspot as well [5], [6]. In addition, subcarrier modulation identification provide important basis for the multiuser and the jamming technology [7], [8], especially in the military application. Lots of research on the identification of non-cooperative underwater acoustic communication has been carried out in the literature. However, the identification of OFDM is mainly focused on the identification of single carrier and multi carrier, and the identification of subcarrier modulation is given less importance.

Various authors have discussed the subcarrier modulation identification of OFDM in the literature and have proposed many schemes, however, most of the application scenarios are focused on the radio. For instance, in [9], the maximum likelihood method is used to identify the subcarrier modulation in the Gaussian channel, but this method is not suitable for the multipath channel. In [10], maximum a posteriori probability (MAP) combined with maximum likelihood is used to identify subcarrier modulation in the fading channel, however, its performance is poor at low SNR. In [11], a high-order cumulant combined with Bayesian is proposed to identify subcarrier modulation, but this method can be identified only when the probability density function (PDF) of the transmitted signal is known, and hence limits its performance in practical use. Although significant results are achieved by the above mentioned schemes, however, still research work is needed to validate these schemes practically. In the aspect of modulation identification in MPSK and multiple quadrature amplitude modulation(MQAM), the identification method based on likelihood is very popular, and shows significant performance when the CIR and noise power are known [12]. However, this identification method draws researcher’s attention when the CIR and noise power are unknown in the multipath fading channel.In [13], [14], EM-QHLRT is proposed for BPSK and QPSK signals. The CIR and noise power are estimated by EM, the QHLRT is used to calculate the likelihood value, and the modulation is identified by comparing the likelihood value. EM-QHLRT effectively improved the identification performance and provided a new idea of modulation identification in fading channel. However, the performance of the channel estimation based on EM is severely affected by the multipath channel.

Combining the characteristics of underwater acoustic OFDM, we propose a novel method of subcarrier modulation identification based on EM-Block-QHLRT, and provide the following contributions in this regard.

  • A novel method of subcarrier modulation identification based on QHLRT is proposed when the CIR and noise power are unknown. After removing cyclic prefix and FFT, the received signal is modeled as Gaussian mixture models (GMM), and the EM-QHLRT is used to identify subcarrier modulation;

  • We proposed the channel estimation strategy of EM-Block to improve the reliability of channel estimation. It is because of the fact that in the multi-carrier of OFDM system, the channels experienced by different subcarriers are different, while the CIR in every subcarrier calculated by EM is the same;

  • We performed simulation analyzing the influence of iteration times, the length of symbols and the number of blocks of the proposed EM-Block-QHLRT method, and then compared the performance of ALRT-UB, EM-QHLRT and EM-Block-QHLRT;

  • We also validated the proposed EM-Block-QHLRT method by using sea trial data.

Section snippets

System model

The difference between underwater acoustic OFDM and single carrier signal indicates that the N-point IFFT transform at transmitter divides the underwater acoustic channel into N parallel subcarrier channels. The OFDM signal generated by the N-point IFFT transform at transmitter can be expressed as:xn,l=k=0N-1cn,kej2πkl/N

Where cn,k is the kth subcarrier modulation signal of the nth OFDM symbol mapped by subcarrier modulation, xn,l is the data on the lth subcarrier of the nth OFDM symbol. Hence,

Parameter setting

In order to verify the validity of the proposed EM-Block-QHLRT method, an underwater acoustic sparse channel is generated for simulation, using the method used in [20] for channel generation. We set the number of paths to 7, and the time delay of adjacent paths follows the exponential distribution using the mean value of 3 ms and the average multipath delay of 21 ms, while the amplitudes are Rayleigh distributed with the average power decreasing exponentially with delay. The noise included in

Experimental results

In this section, we provide the experimental work in Hainan Lingshui, verifying practically the validity of our proposed EM-Block-QHLRT method. In the experiment, we hanged underwater acoustic transducer and hydrophone with two ships. One ship was used as the target node receiving the signal, and the other ship was used as the mobile node transmitting the signals at different distances. The water depth at the experimental site was 60–70 m, and the sea condition was well, however, there was a

Conclusion

A novel EM-Block-QHLRT subcarrier modulation identification method is proposed for the underwater acoustic multipath channel, using the characteristics of multicarrier OFDM, the performance of blind channel estimation is greatly improved. The simulation results showed that the proposed EM-Block-QHLRT method showed better performance than EM-QHLRT method, in terms of improved accuracy. The identification of underwater acoustic OFDM using QPSK subcarrier modulation is verified by sea trial data.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Key R&D Program of China (Grant Nos. 2018YFC0308500), National Natural Science Foundation of China (Grant Nos. 61771152, 11974090, 11774074 and 11704090), the Natural Science Foundation of Heilongjiang Province of China (Grant No. YQ2019F002). The authors wish to thank all participants of the research.

References (20)

  • Gang Qiao et al.

    A low-complexity orthogonal matching pursuit based channel estimation method for time-varying underwater acoustic ofdm systems

    Appl Acoust

    (2019)
  • Yuanyuan Ma et al.

    Modulation format identification based on constellation diagrams in adaptive optical ofdm systems

    Optics Commun

    (2019)
  • Peng Chen et al.

    Joint channel estimation and impulsive noise mitigation in underwater acoustic ofdm communication systems

    IEEE Trans Wireless Commun

    (2017)
  • Lu. Ma et al.

    Superposition coding for downlink underwater acoustic ofdm

    IEEE J Oceanic Eng

    (2017)
  • Jun Tao

    Dft-precoded mimo ofdm underwater acoustic communications

    IEEE J Oceanic Eng

    (2018)
  • Helmut Bolcskei et al.

    Blind channel identification and equalization in ofdm-based multiantenna systems

    IEEE Trans Signal Process

    (2002)
  • Cheong Yui Wong et al.

    Multiuser ofdm with adaptive subcarrier, bit, and power allocation

    IEEE J Sel Areas Commun

    (1999)
  • Jing Mao et al.

    Subcarrier and power allocation for multiuser mimo-ofdm systems with various detectors

    Ksii Trans Internet Inform Syst

    (2017)
  • Yucek T, Huseyin Arslan. A novel sub-optimum maximum-likelihood modulation classification algorithm for adaptive ofdm...
  • L. Haring et al.

    Efficient modulation classification for adaptive wireless ofdm systems in tdd mode

There are more references available in the full text version of this article.

Cited by (15)

  • Categorization of liquid-liquid counter-current two-phase flow-structures in 11 mm ID tube at different conduit-orientations

    2022, International Journal of Multiphase Flow
    Citation Excerpt :

    All these analyses are made by ‘MatLab coding’ using the ‘MATLAB R2016b’ version. A computational intelligence-based categorizer inspired by the EM algorithm (Zhang et al., 2017; Ma et al., 2018; Fang et al., 2021; Gul et al., 2021; Smyth, 2021) is developed to categorize different flow-structures automatically. The statistical parameters extracted from the objective signatures are used as the inputs to this methodology.

  • OTFS underwater acoustic communications based on passive time reversal

    2022, Applied Acoustics
    Citation Excerpt :

    The challenge is more serious for high-rate UWA communications. Orthogonal Frequency Division Multiplexing (OFDM) has widely applied in high-rate UWA communication due to its robustness against long multipath spread channels as well as low complexity [2–6]. The key advantage of OFDM is that the symbols are transmitted in the orthogonal subchannel without interference.

  • Modulation recognition based on deep learning network

    2023, ACM International Conference Proceeding Series
View all citing articles on Scopus
View full text