Elsevier

Applied Acoustics

Volume 175, April 2021, 107687
Applied Acoustics

Self-interference channel modeling for in-band full-duplex underwater acoustic modem

https://doi.org/10.1016/j.apacoust.2020.107687Get rights and content

Abstract

In-band full-duplex (IBFD) underwater acoustic (UWA) communication can significantly improve the throughput of UWA communication networks. In this paper, we focus on the self-interference (SI) channel modeling which is essential for SI cancellation in an IBFD modem. The SI consists of direct self-loop interference (SLI) and self multi-path interference (SMI) due to reflections from water surface and bottom. Therefore, we first propose a simplified finite element model for SLI in an IBFD UWA modem. Then we model the underwater vertical channel to obtain the SMI path loss and propagation delays. Simulation results show that the SLI signal is composed of diffraction and scattering components, and it is greatly affected by the modem housing material and shape. The SLI channel has a long (several tens of ms) impulse response. To verify the proposed model, based on the IBFD UWA communication modem developed by our team, we conducted a lake experiment in December 2019 at Qiandao Lake in Hangzhou, China. The simulated results match well with the experimental results in time/frequency features and transmission loss. This study reveals the complexity of SI channel in IBFD UWA communication.

Introduction

Underwater acoustic (UWA) communication technology has been widely studied and applied in many fields, such as underwater sensor networks, observation of marine environment, oceanographic engineering construction, etc. [1], [2]. Due to the narrow available frequency bandwidth and complex UWA propagation, the spectral efficiency of UWA communication systems is limited [3], [4], [5]. Full-duplex (FD) communication technology was introduced to improve the spectral efficiency of radio communication systems [6], [7], [8], and it can also be used for UWA communication systems.

A series of research work has been devoted to exploring the feasibility of FD technology in UWA communication systems [9], [10], especially the in-band full-duplex (IBFD) technology as it can double the utilization of frequency band and greatly improve the performance of UWA communication networks [11], [12], [13], [14], [15], [16]. In general, the research has been focused on self-interference (SI) cancellation, which is the basis and major challenge of FD communications, and can be implemented as analog SI cancellation and digital SI cancellation. The performance of SI cancellation can be improved with more accurate estimates of the SI channel [11], [12], [13]. Therefore, it is of great significance to study the SI signal and SI propagation channel. In [13], a sparse adaptive constraint algorithm for estimation of the SI channel and power amplifier (PA) nonlinearity is proposed. An improved maximum likelihood (ML) algorithm for the SI channel estimation is proposed in [14], which introduces a penalty that favors sparsity in the cost function to obtain better SI channel estimates. In [11], the SI channel estimates are obtained by using the recursive least-squares (RLS) algorithm with dichotomous coordinate descent (DCD) iterations, which achieves 69 dB SI cancellation performance with the PA output being used as the regressor in the adaptive filter to deal with the non-linear distortions.

It should be noted that the SI signal between transducer and receiver is composed of self-loop interference (SLI) and other self-multi-path interference (SMI). The SLI component propagated through the modem housing is much stronger than the SMI component caused by reflections from the seabed and sea surface. A hybrid design proposed in [12] includes analog, digital cancellation and directional transmission. It can be used for the SLI cancellation in the deep ocean environment. Real sea measurements of SI for FD UWA communication systems are presented in [17], which demonstrate that the SMI can last more than 1 s in shallow water environments.

In practice, the SI cancellation algorithms will run in a communication modem like the one described in [9], so it is essential to consider the influence of the equipment (housing) on the SLI signal. The SMI channel can be modeled by combining some empirical formulas and models [3], [18]. In contrast, it is hard to describe and model the SLI channel realistically at sound propagation distances of only tens of centimeters. Furthermore, the SLI acoustic channel is different from the SLI in IBFD radio channel [19], [20]. The scattering component [21] caused by the IBFD UWA modem housing vibration will also be received by the near-end receiver in the form of interference. With a high transmission power, the far-end signal cannot fit within the limited dynamic range of analog to digital converter (ADC). This requires an analog SI interference cancellation to enable the ADC to convert the far-end signal and further cancel the residual SI by a digital SI canceler [22]. Meanwhile, if some prior information about the SLI channel can be obtained, the complexity of analog interference cancellation can be reduced by digitally assisted analog interference cancellation [23]. Therefore, it is very important to model the SI channel, especially the SLI channel to obtain the prior information.

Hence, in order to better understand the SI channel in practice, especially characteristics of the SLI channel, such as the channel impulse response (CIR), in this work, we develop models for the SLI and SMI channels. First, to focus only on the SLI characteristics and channel modeling, the short distance sound propagation is simulated in infinite space without any interference from multipath components caused by boundaries. We establish a simplified finite element model of an IBFD UWA modem to simulate the sound propagation from the transmitter to the near-end receiver. Then we model the underwater vertical channel to obtain the SMI path loss and arrival time. The simulation results are verified in a lake experiment.

This paper is structured as follows: In Section 2, we describe the modeling of SLI and SMI, the finite element model and parameter configuration. Analysis of the SLI signal is given in Section 3. In Section 4, the simulation results are verified by experiments. Finally, some conclusions and discussions concerning the use of IBFD UWA communication systems in practice are provided in Section 5.

Section snippets

Modeling method and parameter configuration

Normally, the conventional UWA communication modem can be roughly divided into two parts: the transceiver transducer and the housing. However, for the IBFD UWA communication modem, in order to transmit and receive signals at the same time, it is composed of three parts: transmitting transducer, receiving transducer and housing as shown in the left side of Fig. 1. The housing contains the digital, analog circuit boards and battery packs. The housing between the transmitting end and the receiving

Analysis of simulation results

In this section, we use the simplified finite element model to simulate the sound propagation process and obtain the SLI at the near-end receiver. To investigate the influence of the housing on the SLI signal, we also simulate the short-range propagation without the housing for comparison.

Lake experiment

The lake experimental setup is shown in Fig. 7. It was conducted in the Qiandao Lake, Hangzhou China. It was carried out on a large floating platform more than 100 m away from the shore. The depth of the lake water is about 38 m. In order to verify the SLI and SMI channel, we put the hydrophone at a depth of 14 meters, so that the receiving end was not affected by the reflection of lake surface and bottom in a certain observation time (about 18 ms which is much longer than the duration of the

Conclusions and discussions

In this study, we used a simplified finite element model of the IBFD UWA communication modem to simulate the influence of the housing on the SI at near-end receiver. The simulation results show that the SLI signal received by the near-end receiver contains diffraction component and scattering component, and the scattering component is more intensive. We also modeled the SMI channel and obtained the path loss and arrival time of different taps. The simulation results are verified in the lake

CRediT authorship contribution statement

Yunjiang Zhao: Conceptualization, Methodology, Software, Formal analysis, Validation, Data curation, Investigation, Writing - original draft, Writing - review & editing. Gang Qiao: Conceptualization, Resources, Formal analysis, Validation, Project administration, Funding acquisition, Investigation. Songzuo Liu: Software, Resources, Validation, Data curation, Writing - original draft, Project administration, Investigation. Yuriy Zakharov: Writing - original draft, Visualization, Supervision.

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 under (Grant Nos. 61771152, 11974090, 11774074 and 11704090), the Natural Science Foundation of Heilongjiang Province of China (Grant No. YQ2019F002). The Key Laboratory of Acoustic Environment, Institute of Acoustics, Chinese Academy of Sciences (Grant No. SHHJKFKT-1801) and Innovation Special Zone of National Defense Science and Technology.

The work of Y.

References (32)

  • G. Qiao et al.

    Full-duplex, multi-user and parameter reconfigurable underwater acoustic communication modem

  • L. Shen et al.

    Digital self-interference cancellation for full-duplex underwater acoustic systems

    IEEE Trans Circuits Syst II Express Briefs

    (2019)
  • L. Li et al.

    Interference cancellation in in-band full-duplex underwater acoustic systems

  • G. Qiao et al.

    Self-interference channel estimation algorithm based on maximum-likelihood estimator in in-band full-duplex underwater acoustic communication system

    IEEE Access

    (2018)
  • G. Qiao et al.

    Digital self-interference cancellation for asynchronous in-band full-duplex underwater acoustic communication

    Sensors

    (2018)
  • F. Qu et al.

    In-band full-duplex communications for underwater acoustic networks

    IEEE Network

    (2017)
  • View full text