Using BLE Signals to Estimate Objective and Subjective Crowdedness Levels on Fixed-Route Buses
Accurately estimating the crowdedness inside a fixed-route bus is essential for improving transportation system efficiency and enhancing passenger comfort. While methods using cameras or sensors installed at bus entrances to count passengers have been proposed, these methods present challenges in te...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10955370/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Accurately estimating the crowdedness inside a fixed-route bus is essential for improving transportation system efficiency and enhancing passenger comfort. While methods using cameras or sensors installed at bus entrances to count passengers have been proposed, these methods present challenges in terms of passenger privacy, installation costs, and placement. These approaches typically use the number of passengers as an objective indicator to evaluate crowdedness. However, even with the same number of passengers, the subjective crowdedness level experienced by each passenger can vary. Thus, it is important to estimate the objective crowdedness and the subjective crowdedness level perceived by bus passengers. In this study, we developed a method for estimating objective and subjective crowdedness levels using only Bluetooth Low Energy (BLE) information collected by the existing bus location tracking system installed in fixed-route buses to reduce privacy and installation costs. Specifically, Bluetooth device (BD) addresses obtained from BLE scans are filtered based on occurrence frequency and average RSSI to distinguish between passenger and surrounding BD addresses. The number of passenger BD addresses, along with their differences and rates of change, are used as features to estimate the number of passengers and the subjective crowdedness level using machine learning models. An experiment to evaluate the BLE method produced an accuracy of 0.653 for the objective crowdedness level (number of passengers) and 0.513 for the subjective crowdedness level, indicating that BLE signal information can capture the general trend of objective and subjective crowdedness. |
|---|---|
| ISSN: | 2169-3536 |