Accurate localization is critical for Internet of Things (IoT) applications. Using hop loss in DV-Hop-based algorithms is a promising approach. Nevertheless, challenges lie in overcoming the computational complexity caused by re-calculating the predicted hop-counts, and how to further optimize the modeling for better accuracy. In this paper, a novel hop loss modeling, distance-based connectivity consistency (DCC), is proposed. By focusing on the first order connectivity, DCC avoids computing predicted hop-counts, and significantly reduces the time complexity. We also provide a proof to theoretically guarantee that this design achieves a full coverage of all hop errors. In addition, by computing a continuous loss function instead of the discrete hop-count errors, DCC further improves the localization accuracy. In the evaluations, DCC demonstrates notable improvements in accuracy over other highly regarded algorithms, and reduces 30% to 40% total computation time compared with the baseline algorithm using hop loss.

翻译：精准定位对物联网应用至关重要。在基于DV-Hop的算法中利用跳数损失是一种前景广阔的方法。然而，其挑战在于克服因重新计算预测跳数而带来的计算复杂度，以及如何进一步优化建模以提升精度。本文提出了一种新颖的跳数损失建模方法——基于距离的连通性一致性。该方法聚焦于一阶连通性，避免了预测跳数的计算，显著降低了时间复杂度。我们同时提供了理论证明，确保该设计能够完全覆盖所有跳数误差。此外，通过计算连续损失函数而非离散的跳数误差，该方法进一步提升了定位精度。在评估实验中，相较于其他备受关注的算法，该方法在精度上表现出显著提升，并且相比采用跳数损失的基线算法，总计算时间减少了30%至40%。