Enhanced Distance Vector Hop Localization Algorithm based on Hop Threshold and Weighted Matrix for Wireless Sensor Networks

Abstract

The Distance Vector Hop (DV-Hop) algorithm is a frequently used localization method, which is widely applied in location-based applications for its implement-efficient. However, it has poor localization accuracy, especially in complex unevenly distributed network environment. This study proposed an enhanced range-free localization scheme based on hop threshold and weighted matrix to address this issue, named as TWDV-Hop. First, we build considerable experiments to analyze the distributed law between hop count and average hop size error. Large hop count will be cut out to optimize average hop size based on analysis results, since it is the main reason that led to large location error for its inaccuracy. Then, weighted matrix is introduced to instead of basic least squares in the third phase to narrow location error brought by non-linear equation. Finally, extensive experiments were conducted for several evaluation metrics, as localization accuracy, energy consumption cost under effected parameters in terms of anchor node density, communication range etc. Simulation results demonstrated the TWDV-Hop algorithm had outstanding performance in accuracy and energy consumption. The localization error is decreased more than 60%, when compared with latest new literatures. Especially, the maximum reduction localization error reached up to 75%. Moreover, the average localization error is lower 3.5m, which can meet location-based application requirements at a certain level.