Abstract
Conference Title: 2016 IEEE International Conference on Ubiquitous Wireless Broadband (ICUWB) Conference Start Date: 2016, Oct. 16 Conference End Date: 2016, Oct. 19 Conference Location: Nanjing, China The blocking of line-of-sight between anchors and targets by distributed (i.e., non-point) obstacles in an environment can create blind-spots in a localization network if there are an insufficient number of unblocked anchors. The spatial randomness of the obstacle and anchor locations makes it difficult to characterize the blind-spot probability of a network. In this paper, we use tools from stochastic geometry to characterize this randomness. In particular, a homogeneous Poisson point process is used to model the anchor locations and a germ-grain model is employed to represent obstacle locations and shapes. Unlike previous works, which usually assume independent blocking of anchors, we develop tools to handle correlated blocking. Specifically, we use a mixture distribution to approximate the variance of the unshadowed area, which in turn, is used to approximate the blind-spot probability.