Design, Analysis, and Field Testing of an Innovative Drone-Assisted Zero-Configuration Localization Framework for Wireless Sensor Networks

Abstract

In wireless sensor networks, there are usually a small number of position-aware nodes, referred to as anchors. Every other node, called target node (or sensor), can estimate its distances to surrounding anchors according to the received signal strength. Then, trilateration can be employed for self localization. This system is low-cost, simple, and thus popular, but it suffers from two major drawbacks. The first problem is the requirement for realtime channel parameters, which should be estimated and transmitted to target sensors by a central node. Another problem arises from the fact that positioning accuracy is dependent on number of anchors, but anchor density is quite limited for economic consideration in the conventional localization systems. To solve these problem, we proposed in this paper, a new localization framework for wireless sensor networks, which employs a mobile node to serve as virtual anchors. This new framework has two advantages over the conventional one: 1) Offline measurements are totally unnecessary; 2) the number of virtual anchors can be vastly increased with negligible cost, leading to very high positioning accuracy. To justify the superiority of this new framework, extensive analysis and simulations were conducted. Besides, we implemented it on a drone and did field experiments, which showed very promising results.