Interpolation of Sensory Data in the Presence of Obstacles

Due to the inherent discrete nature of sensing, interpolation is a key activity in sensing. Interpolation in urban computing faces an unprecedented inference issue that was not present in large scale traditional sensing environments. Obstacles such as buildings and walls are prevalent in urban sensing, and it is important to consider them in urban sensing interpolation. This paper introduces an obstacle handling interpolation in urban sensing where obstacles are present. Experimental results demonstrate that our proposed method outperforms traditional well-known interpolation methods, and test statistics verifies our method significantly improves performance at the expense of small amount of extra time.