Transfer Learning for Wireless Fingerprinting Localization Based on Optimal Transport
Wireless fingerprinting localization (FL) systems identify locations by building radio fingerprint maps, aiming to provide satisfactory location solutions for the complex environment. However, the radio map is easy to change, and the cost of building a new one is high. One research focus is to trans...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/23/6994 |
| _version_ | 1827700656038739968 |
|---|---|
| author | Siqi Bai Yongjie Luo Qun Wan |
| author_facet | Siqi Bai Yongjie Luo Qun Wan |
| author_sort | Siqi Bai |
| collection | DOAJ |
| description | Wireless fingerprinting localization (FL) systems identify locations by building radio fingerprint maps, aiming to provide satisfactory location solutions for the complex environment. However, the radio map is easy to change, and the cost of building a new one is high. One research focus is to transfer knowledge from the old radio maps to a new one. Feature-based transfer learning methods help by mapping the source fingerprint and the target fingerprint to a common hidden domain, then minimize the maximum mean difference (MMD) distance between the empirical distributions in the latent domain. In this paper, the optimal transport (OT)-based transfer learning is adopted to directly map the fingerprint from the source domain to the target domain by minimizing the Wasserstein distance so that the data distribution of the two domains can be better matched and the positioning performance in the target domain is improved. Two channel-models are used to simulate the transfer scenarios, and the public measured data test further verifies that the transfer learning based on OT has better accuracy and performance when the radio map changes in FL, indicating the importance of the method in this field. |
| first_indexed | 2024-03-10T14:17:05Z |
| format | Article |
| id | doaj.art-fa90c807af9f45d49d2f061ce4fcb409 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T14:17:05Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-fa90c807af9f45d49d2f061ce4fcb4092023-11-20T23:45:21ZengMDPI AGSensors1424-82202020-12-012023699410.3390/s20236994Transfer Learning for Wireless Fingerprinting Localization Based on Optimal TransportSiqi Bai0Yongjie Luo1Qun Wan2School of Information and Communication Engineering, University of Electronic Science and Technology of China, No.2006, Xiyuan Ave, West Hi-Tech Zone, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, No.2006, Xiyuan Ave, West Hi-Tech Zone, Chengdu 611731, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, No.2006, Xiyuan Ave, West Hi-Tech Zone, Chengdu 611731, ChinaWireless fingerprinting localization (FL) systems identify locations by building radio fingerprint maps, aiming to provide satisfactory location solutions for the complex environment. However, the radio map is easy to change, and the cost of building a new one is high. One research focus is to transfer knowledge from the old radio maps to a new one. Feature-based transfer learning methods help by mapping the source fingerprint and the target fingerprint to a common hidden domain, then minimize the maximum mean difference (MMD) distance between the empirical distributions in the latent domain. In this paper, the optimal transport (OT)-based transfer learning is adopted to directly map the fingerprint from the source domain to the target domain by minimizing the Wasserstein distance so that the data distribution of the two domains can be better matched and the positioning performance in the target domain is improved. Two channel-models are used to simulate the transfer scenarios, and the public measured data test further verifies that the transfer learning based on OT has better accuracy and performance when the radio map changes in FL, indicating the importance of the method in this field.https://www.mdpi.com/1424-8220/20/23/6994fingerprinting localizationtransfer learningoptimal transportindoor positioningadaptive radio map |
| spellingShingle | Siqi Bai Yongjie Luo Qun Wan Transfer Learning for Wireless Fingerprinting Localization Based on Optimal Transport Sensors fingerprinting localization transfer learning optimal transport indoor positioning adaptive radio map |
| title | Transfer Learning for Wireless Fingerprinting Localization Based on Optimal Transport |
| title_full | Transfer Learning for Wireless Fingerprinting Localization Based on Optimal Transport |
| title_fullStr | Transfer Learning for Wireless Fingerprinting Localization Based on Optimal Transport |
| title_full_unstemmed | Transfer Learning for Wireless Fingerprinting Localization Based on Optimal Transport |
| title_short | Transfer Learning for Wireless Fingerprinting Localization Based on Optimal Transport |
| title_sort | transfer learning for wireless fingerprinting localization based on optimal transport |
| topic | fingerprinting localization transfer learning optimal transport indoor positioning adaptive radio map |
| url | https://www.mdpi.com/1424-8220/20/23/6994 |
| work_keys_str_mv | AT siqibai transferlearningforwirelessfingerprintinglocalizationbasedonoptimaltransport AT yongjieluo transferlearningforwirelessfingerprintinglocalizationbasedonoptimaltransport AT qunwan transferlearningforwirelessfingerprintinglocalizationbasedonoptimaltransport |