Scholarlink

<p>This paper addresses the problem of distributed estimation in sensor networks with uncertain inter-node channel gains. Such uncertainties, if not properly accounted for, can significantly degrade estimation accuracy. To mitigate this issue, the channel gain errors are modeled as additive Gaussian perturbations, and closed-form Maximum A Posteriori (MAP) estimators are derived. Leveraging these estimates, two improved version of Diffusion Least Mean Square (DLMS) algorithm are developed: one that computes optimal combination coefficients to minimize uncertainty-induced disturbances, and another that explicitly minimizes the mean uncertain disturbance. Simulation results demonstrate that the proposed methods significantly outperform conventional approaches, particularly under severe channel uncertainty.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Distributed estimation in presence of uncertain channel gain for a sensor network

  • 期刊论文

  • 发表日期：

    2025年8月4日

查看全文
  • Hadi Zayyani,
  • Mahdi Shamsi,
  • Hasan Abu Hilal,
  • Mohammad Salman

摘要

This paper addresses the problem of distributed estimation in sensor networks with uncertain inter-node channel gains. Such uncertainties, if not properly accounted for, can significantly degrade estimation accuracy. To mitigate this issue, the channel gain errors are modeled as additive Gaussian perturbations, and closed-form Maximum A Posteriori (MAP) estimators are derived. Leveraging these estimates, two improved version of Diffusion Least Mean Square (DLMS) algorithm are developed: one that computes optimal combination coefficients to minimize uncertainty-induced disturbances, and another that explicitly minimizes the mean uncertain disturbance. Simulation results demonstrate that the proposed methods significantly outperform conventional approaches, particularly under severe channel uncertainty.