A scalable deep neural network architecture for multi-building and multi-floor indoor localization based on Wi-Fi fingerprinting

Kyeong Soo Kim, Sanghyuk Lee, Kaizhu Huang
  • Big Data Analytics, April 2018, Springer Science + Business Media
  • DOI: 10.1186/s41044-018-0031-2

Deep-neural-network-based multi-building and multi-floor indoor localization scheme

What is it about?

We use just a single deep neural network for building, floor, and floor-level location classification and coordinates estimation, which can provide near state-of-the-art localization performance.

Why is it important?

The proposed scheme makes much simpler the management of Wi-Fi fingerprint databases and related training and thereby provides good scalability compared to existing schemes.

Perspectives

Dr Kyeong Soo Kim (Author)
Xi'an Jiaotong-Liverpool University

This paper opens a new research area for scalable multi-building and multi-floor indoor localization enabled by deep neural networks based on Wi-Fi fingerprinting. Note that most of existing works in this area have been focusing on the indoor localization in a single, connected space.

Read Publication

http://dx.doi.org/10.1186/s41044-018-0031-2

The following have contributed to this page: Dr Kyeong Soo Kim