Fig. 18: Percentage of localization error vs. Total Number of Nodes with 0-30% ranging error.
As node density per unit area increases with an increase in the total number of nodes, it makes the unknown nodes closer to more anchor nodes. This way the lesser hop counts determines the distance \(d_{i}\) between the anchor node and the unknown node with less error is governed by equation (2) and (23). The more appropriate distance estimation\(d_{i}\) contributes a better location estimation for the DV-Hop algorithm and IDV. Similarly, ODR improves due to more anchor nodes fall in the minimum distant anchor nodes set \({}^{\prime}K^{\prime}\) and hence the centroidCN can indicate a probable location for an intended unknown node with more precision.
The Experiment 3 reveals that the proposed model ODR is better than DV-Hop algorithm and IDV by approximately 7% and 4% respectively in the ideal case as shown by Fig. 15 while the improvement is approximately 10%, 14%, 20% and 8%, 10%, 12% in comparison to DV-Hop algorithm and IDV respectively by considering the ranging error slabs of 0-10%, 0- 20%, and 0-30% respectively as presented through Fig. 16, 17, and 18.