Human-to-Human Position Estimation System Using RSSI in Outdoor Environment.

Area of Science:

• Engineering
• Computer Science
• Robotics

Background:

• Traffic accident prevention and collision avoidance are critical safety concerns.
• Accurate positioning in non-line-of-sight (NLOS) environments is challenging with conventional methods like GPS and Received Signal Strength Indication (RSSI).
• Existing RSSI methods often require multiple receivers, leading to reduced accuracy due to environmental obstructions.

Purpose of the Study:

• To propose an improved system for accurate position estimation in NLOS areas.
• To address the limitations of conventional RSSI-based positioning systems.
• To enable the estimation of unspecified transmitter positions using Bluetooth Low Energy (BLE) advertising signals.

Main Methods:

• Developed a novel position estimation system utilizing RSSI MAP simulation and a particle filter.
• Employed BLE peripheral/central functions for transmitters and receivers, leveraging advertising radio waves.
• Validated the system through both computer simulations and real-world experiments.

Main Results:

• Simulations demonstrated a high precision with an average distance error of 1.6 meters.
• Experiments in actual environments yielded an average distance error of 3.3 meters.
• The system maintained comparable accuracy (4.5m error) for both stationary and moving transmitters/receivers in outdoor NLOS conditions.

Conclusions:

• The proposed BLE-based RSSI MAP and particle filter system offers accurate positioning in outdoor NLOS environments.
• The method provides a viable solution for collision avoidance systems by detecting objects in obstructed areas.
• The system's accuracy is comparable to conventional methods but extends functionality to NLOS scenarios.