Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Quasi-deterministic Channel Model for Underwater Acoustic Communication Systems

Yuxuan Yang, Yilin Ma, Hengtai Chang, Cheng-Xiang Wang

TL;DR

This work tackles the challenge of modeling non-stationary shallow-water underwater acoustic channels by introducing a quasi-deterministic framework that fuses BELLHOP (deterministic) with a geometry-based stochastic model (GBSM). It classifies multipath components into D-rays (deterministic, including LoS and boundary reflections), R-rays (diffuse surface/boundary scattering), and F-rays (in-water clusters with birth-death dynamics), and derives a composite channel transfer function $H(t,f)=\sqrt{S_\mathrm{D}}H_\mathrm{D}(t,f)+\sqrt{S_\mathrm{R}}H_\mathrm{R}(t,f)+\sqrt{S_\mathrm{F}}H_\mathrm{F}(t,f)$ to capture their combined effects. The paper provides explicit modeling flavors for each ray type, including Doppler treatment and environmental-loss factors, and develops statistical tools such as TF-CF, coherence time, and Doppler PSD to quantify non-stationarity. Simulation results corroborate the model’s ability to reproduce time-varying ACFs, carrier-frequency dependent coherence, and Doppler characteristics under realistic shallow-water conditions. The approach offers a flexible, accurate framework for designing UWA systems and evaluating performance in non-stationary environments.

Abstract

In this paper, a quasi-deterministic (Q-D) model for non-stationary underwater acoustic (UWA) channels is proposed. This model combines the BELLHOP deterministic model and geometry-based stochastic model (GBSM), which provides higher accuracy and flexibility. Different propagation components in shallow water are classified as D-rays, R-rays and F-rays in the proposed model, where D-rays are modeled by BELLHOP while both R-rays and F-rays are modeled by GBSM. Some important channel statistical properties, including time-frequency correlation function (TF-CF), Doppler power spectrum density (PSD), average Doppler shift, and RMS Doppler spread are derived and simulated. Finally, simulation results illustrate the correctness of the proposed model.

A Quasi-deterministic Channel Model for Underwater Acoustic Communication Systems

TL;DR

This work tackles the challenge of modeling non-stationary shallow-water underwater acoustic channels by introducing a quasi-deterministic framework that fuses BELLHOP (deterministic) with a geometry-based stochastic model (GBSM). It classifies multipath components into D-rays (deterministic, including LoS and boundary reflections), R-rays (diffuse surface/boundary scattering), and F-rays (in-water clusters with birth-death dynamics), and derives a composite channel transfer function to capture their combined effects. The paper provides explicit modeling flavors for each ray type, including Doppler treatment and environmental-loss factors, and develops statistical tools such as TF-CF, coherence time, and Doppler PSD to quantify non-stationarity. Simulation results corroborate the model’s ability to reproduce time-varying ACFs, carrier-frequency dependent coherence, and Doppler characteristics under realistic shallow-water conditions. The approach offers a flexible, accurate framework for designing UWA systems and evaluating performance in non-stationary environments.

Abstract

In this paper, a quasi-deterministic (Q-D) model for non-stationary underwater acoustic (UWA) channels is proposed. This model combines the BELLHOP deterministic model and geometry-based stochastic model (GBSM), which provides higher accuracy and flexibility. Different propagation components in shallow water are classified as D-rays, R-rays and F-rays in the proposed model, where D-rays are modeled by BELLHOP while both R-rays and F-rays are modeled by GBSM. Some important channel statistical properties, including time-frequency correlation function (TF-CF), Doppler power spectrum density (PSD), average Doppler shift, and RMS Doppler spread are derived and simulated. Finally, simulation results illustrate the correctness of the proposed model.
Paper Structure (13 sections, 33 equations, 3 figures, 1 table)

This paper contains 13 sections, 33 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. A Quasi-Deterministic Model Combining BELLHOP and GBSM
  3. D-rays Modeling
  4. R-rays Modeling
  5. F-rays Modeling
  6. Doppler Effect
  7. Statistical Properties
  8. TF-CF
  9. Coherence Time
  10. Doppler PSD
  11. Average Doppler shift and RMS Doppler spread
  12. Results and Analysis
  13. Conclusion

Figures (3)

  • Figure 1: Different components of Q-D model
  • Figure 2: Temporal ACFs at different times and carrier frequencies ($K = 1.6, S_\mathrm{D} = 0.4, S_\mathrm{R} = 0.4, S_\mathrm{F} = 0.2$).
  • Figure 3: Doppler PSDs at different times and carrier frequencies ($K = 1.6, S_\mathrm{D} = 0.4, S_\mathrm{R} = 0.4, S_\mathrm{F} = 0.2$).