Space-Time Hierarchical-Graph Based Cooperative Localization in Wireless Sensor Networks

TL;DR

This paper introduces a space-time hierarchical-graph scheme for cooperative localization in wireless sensor networks, improving accuracy and efficiency by layered position estimation, optimized message passing, and energy-saving mechanisms.

Contribution

It proposes a novel hierarchical-graph based method that enhances localization accuracy and reduces computational and energy costs in sensor networks.

Findings

The scheme exhibits fast convergence in localization tasks.

It reduces energy consumption and network traffic.

It outperforms benchmark localization schemes.

Abstract

It has been shown that cooperative localization is capable of improving both the positioning accuracy and coverage in scenarios where the global positioning system (GPS) has a poor performance. However, due to its potentially excessive computational complexity, at the time of writing the application of cooperative localization remains limited in practice. In this paper, we address the efficient cooperative positioning problem in wireless sensor networks. A space-time hierarchical-graph based scheme exhibiting fast convergence is proposed for localizing the agent nodes. In contrast to conventional methods, agent nodes are divided into different layers with the aid of the space-time hierarchical-model and their positions are estimated gradually. In particular, an information propagation rule is conceived upon considering the quality of positional information. According to the rule, the information always propagates from the upper layers to a certain lower layer and the message passing process is further optimized at each layer. Hence, the potential error propagation can be mitigated. Additionally, both position estimation and position broadcasting are carried out by the sensor nodes. Furthermore, a sensor activation mechanism is conceived, which is capable of significantly reducing both the energy consumption and the network traffic overhead incurred by the localization process. The analytical and numerical results provided demonstrate the superiority of our space-time hierarchical-graph based cooperative localization scheme over the benchmarking schemes considered.