The technology of underwater wireless acoustic sensor networks (UWSNs) plays an important role in many commercial and military applications in underwater. In UWSNs, there exist many challenges posed by the specific characteristics of underwater acoustic channel, such as sound propagation speed is a function of depth with unknown parameters. Besides, time synchronization of anchor nodes is a crucial part of the applicable network. Time synchronization of anchor nodes as well as estimation of sound speed parameters considering an isogradient profile is explored here. While the previous efforts do not take into account these factors, this article assumes only one synchronized anchor node in the network and enables joint synchronization of the other anchor nodes and the estimation of the propagation delays between the anchor nodes using a weighted linear least square solution. Then, it uses the estimated propagation delays to estimate the unknown parameters of the sound speed profile using a conventional Gauss-Newton algorithm with approximately three iterations to converge. Validation of the proposed method is done by the numerical simulations and to demonstrate the effectiveness of the proposed method we compare the results with Cramer-Rao bound average.

中文翻译：

---

异步网络中的水下声速剖面参数估计

水下无线声学传感器网络 (UWSN) 技术在许多水下商业和军事应用中发挥着重要作用。在 UWSN 中，水下声学信道的特定特性存在许多挑战，例如声音传播速度是深度的函数，参数未知。此外，锚节点的时间同步是适​​用网络的重要组成部分。此处探讨了锚节点的时间同步以及考虑等梯度分布的声速参数估计。虽然之前的努力没有考虑到这些因素，本文假设网络中只有一个同步锚节点，并启用其他锚节点的联合同步，并使用加权线性最小二乘法估计锚节点之间的传播延迟。然后，它使用估计的传播延迟来估计声速剖面的未知参数，使用传统的高斯-牛顿算法大约三次迭代收敛。所提出方法的验证是通过数值模拟完成的，为了证明所提出方法的有效性，我们将结果与 Cramer-Rao 边界平均值进行了比较。它使用估计的传播延迟来估计声速曲线的未知参数，使用传统的高斯-牛顿算法大约三次迭代收敛。所提出方法的验证是通过数值模拟完成的，为了证明所提出方法的有效性，我们将结果与 Cramer-Rao 边界平均值进行了比较。它使用估计的传播延迟来估计声速曲线的未知参数，使用传统的高斯-牛顿算法大约三次迭代收敛。所提出方法的验证是通过数值模拟完成的，为了证明所提出方法的有效性，我们将结果与 Cramer-Rao 边界平均值进行了比较。