arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1502.01699eess.SYcs.SY
收藏

无线传感器网络协同定位中唯一局部可定位性的图不变量:刚性指数与冗余指数

Graph invariants for unique localizability in cooperative localization of wireless sensor networks: rigidity index and redundancy index

Kirikkale University(基里赫莱大学)

Tolga Eren

AI总结:

本文提出基于图不变量的刚性理论方法,通过定义刚性指数和冗余指数来评估无线传感器网络图结构的刚性和冗余性,以确定协同定位的唯一局部可定位性。

摘要:

刚性理论使我们能够指定无线传感器网络协同定位问题中唯一局部可定位性的条件。本文提出了一种组合刚性方法,通过图不变量来测量图结构的刚性和广义冗余刚性属性。我们定义刚性指数为基于刚性矩阵中的边独立集的图不变量,其值介于0和1之间,表示接近刚性的程度。冗余刚性对于全局刚性是必需的,与图的唯一实现相关。此外,冗余刚性还为网络系统提供了对结构变化(如链路丢失)的刚性鲁棒性。这里,我们给出了更广泛的冗余边定义,称为“广义冗余边”。该冗余定义适用于刚性和非刚性图。接下来,我们定义冗余指数为基于刚性矩阵中广义冗余边的图不变量,其值介于0和1之间,表示图中冗余的百分比。这两个指数使我们能够探索从非刚性到刚性以及从刚性到冗余刚性的转变。通过图例展示了这种方法。从传感器网络的角度来看,这两个指数使我们能够评估传感器感知半径对网络刚性特性的影响,从而检查传感器网络的局部可定位性。通过使用随机几何图进行仿真,我们评估了通过刚性指数和冗余指数确定局部可定位性所需的变化感知半径。

英文摘要:

Rigidity theory enables us to specify the conditions of unique localizability in the cooperative localization problem of wireless sensor networks. This paper presents a combinatorial rigidity approach to measure (i) generic rigidity and (ii) generalized redundant rigidity properties of graph structures through graph invariants for the localization problem in wireless sensor networks. We define the rigidity index as a graph invariant based on independent set of edges in the rigidity matroid. It has a value between 0 and 1, and it indicates how close we are to rigidity. Redundant rigidity is required for global rigidity, which is associated with unique realization of graphs. Moreover, redundant rigidity also provides rigidity robustness in networked systems against structural changes, such as link losses. Here, we give a broader definition of redundant edge that we call the "generalized redundant edge." This definition of redundancy is valid for both rigid and non-rigid graphs. Next, we define the redundancy index as a graph invariant based on generalized redundant edges in the rigidity matroid. It also has a value between 0 and 1, and it indicates the percentage of redundancy in a graph. These two indices allow us to explore the transition from non-rigidity to rigidity and the transition from rigidity to redundant rigidity. Examples on graphs are provided to demonstrate this approach. From a sensor network point of view, these two indices enable us to evaluate the effects of sensing radii of sensors on the rigidity properties of networks, which in turn, allow us to examine the localizability of sensor networks. We evaluate the required changes in sensing radii for localizability by means of the rigidity index and the redundancy index using random geometric graphs in simulations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑