Abstract
Many applications have been identified which require the deployment of large-scale low-power wireless sensor networks. Some of the deployment environments, however, impose harsh operation conditions due to intense cross-technology interference, extreme weather conditions (heavy rainfall, excessive heat, etc.), or rough motion, thereby affecting the quality and predictability of the wireless links the nodes establish. In localization tasks, these conditions often lead to significant errors in estimating the position of target nodes. Motivated by the practical deployments of sensors on the surface of different water bodies, we address the problem of identifying susceptible nodes and robustly estimating their positions. We formulate these tasks as a compressive sensing problem and propose algorithms for both node identification and robust estimation. Additionally, we design an optimal anchor configuration to maximize the robustness of the position estimation task. Our numerical results and comparisons with competitive methods demonstrate that the proposed algorithms achieve both objectives with a modest number of anchors. Since our method relies only on target-to-anchor distances, it is broadly applicable and yields resilient, robust localization.
Abstract (translated)
许多应用场景已确定需要部署大规模低功耗无线传感器网络。然而，一些部署环境由于存在强烈的跨技术干扰、极端天气条件（如暴雨和高温）或剧烈的运动等因素，给节点建立无线链路的质量和可预测性带来了挑战，从而导致定位任务中目标节点位置估计出现显著误差。 受在不同水体表面部署传感器的实际应用启发，我们致力于识别易受影响的节点并稳健地估算其位置。我们将这些问题转化为压缩感知问题，并提出了用于节点识别和稳健估计的算法。此外，我们还设计了一种最优锚点配置方案以最大化定位任务的鲁棒性。我们的数值结果与竞争方法相比表明，所提出的算法能够使用少量锚点实现这两个目标。 由于本方法仅依赖于目标到锚点的距离信息，它具有广泛适用性和提供稳健、精确的定位能力的特点。
URL
https://arxiv.org/abs/2507.03856
