Parallel kd-Tree Based Approach for Computing the Prediction Horizon Using Wolf’s Method

Figure 3

The kd-tree that represents the partitions in Figure 2. The internal nodes have the cut dimension and the cut value for each partition. For each level, all the points contained in the left subtree have values less than or equal to the cut value in the cut dimension; all the points contained in the right subtree have values greater than the cut value in the cut dimension. The external nodes, or buckets, store the points in the resulting hyperrectangles of the partitions.