Distance Estimation Between Wireless Sensor Nodes Using RSSI and CSI with Bounded-Error Estimation and Theory of Evidence for a Landslide Monitoring System

The wireless sensor network (WSN) deployed in a landslide-prone area is equipped with sensors capturing environmental parameters, e.g., rainfall, moisture, etc. Among them, the ground displacement rate is essential. However, measuring it requires additional sensors such as an extensometer and their installation by the specialist. This study aims to utilize some information available in the WSN without addition to do the task. The received signal strength indicator (RSSI) and the channel state information (CSI) are used in this work since they relate to the distance between nodes. However, both are sensitive to environmental noise, resulting in an inaccurately estimated displacement. Therefore, a new information fusion framework based on bounded-error estimation is proposed to improve the accuracy. Under this framework, the theory of evidence is used to model variable uncertainties. Experimental results showed that the accuracy of the proposed method was increased by at least approximately 4.5% compared to conventional ones.

I. Introduction

A landslide is one of the most dangerous natural hazards that can occur globally. It happens when the ground changes its condition from stable to unstable, including a wide range of ground movements, such as rockfall, slope failure, and debris flow. Landslides cause significant impacts to our socioeconomic, for example, property damage, death, and injury [1]. Therefore, monitoring of landslides in risk areas is essential in the prevention and mitigation of those losses. Among various approaches, the monitoring system based on a wireless sensor network is considered a promising approach [2].