Iterative Vector-based Localization in a Large Heterogeneous Sensor Network

This paper proposes a novel iterative vector-based localization method in a large heterogeneous sensor network, where a subset of nodes possesses the capability to measure both distance and angle information, while the others are only limited to distance measurements. Unlike conventional vector-based positioning methods that assume all nodes can measure both distance and angle, our approach tackles a more realistic scenario where some nodes are limited to distance-only measurements. To address the challenges of the node localization in a heterogeneous sensor network, the proposed positioning method calculates vector information between the nodes that are not directly communicated and aligns it with a reference coordinate. In addition, the proposed method employs an iterative calculation, such as least-squares minimization, thereby achieving high positioning accuracy. Simulation results demonstrate that the proposed positioning method outperforms the conventional distance-based positioning method in environments with low angle measurement errors, exhibiting up to 44% higher positioning accuracy. Furthermore, the proposed positioning method shows 24% higher positioning accuracy compared to the conventional vector-based positioning method.