Decentralized adaptive indoor positioning protocol using Bluetooth Low Energy
| Autor: | Henry C. B. Chan, Yik Him Ho |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Computer Networks and Communications
Computer science business.industry computer.internet_protocol Node (networking) Real-time computing Process (computing) 020206 networking & telecommunications 02 engineering and technology law.invention Bluetooth Radio propagation Broadcasting (networking) law 0202 electrical engineering electronic engineering information engineering Wireless 020201 artificial intelligence & image processing business Protocol (object-oriented programming) computer Bluetooth Low Energy |
| Zdroj: | Computer Communications. 159:231-244 |
| ISSN: | 0140-3664 |
| DOI: | 10.1016/j.comcom.2020.04.041 |
| Popis: | Previous indoor positioning research has mainly been focused on using Wi-Fi and RFID. In recent years, researchers began to study using Bluetooth 4.0 and Bluetooth Low Energy (BLE) for indoor positioning purposes. In general, positioning techniques based on received signal strength indicator (RSSI), such as signal propagation and fingerprint, are commonly used in wireless/mobile networks. These techniques have certain limitations and tradeoff in terms of accuracy, ease of implementation and practical application/deployment. For example, both methods require a training process before deployment. In this paper, we present a decentralized BLE-based positioning protocol that does not require training before deployment. The training process can automatically be done on the fly by the anchor nodes. While the anchor nodes are broadcasting, they also scan for signals emitted by other anchors. This collaborative communication process exchanges location information and signal strength measurements between each anchor. This process builds a signal-to-distance reference list for the target node to estimate physical distance in a more accurate way. Experimentation in a real indoor environment shows that the proposed collaborative positioning method can achieve an error of 1.5 meters on average. This is generally applicable for most indoor positioning applications for locating people. Furthermore, its implementation is simple and practical, because it does not require training before positioning estimation and is adaptive to environmental changes. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect