loading page

Personalized Federated Learning on NLOS Acoustic Signal Classification
  • +3
  • Hucheng Wang,
  • Suo Qiu,
  • Jingjing Wang,
  • Lei Zhang,
  • Zhi Wang,
  • Xiaonan Luo
Hucheng Wang
Guilin University of Electronic Technology

Corresponding Author:[email protected]

Author Profile
Suo Qiu
Zhejiang University
Author Profile
Jingjing Wang
Kyungpook National University
Author Profile
Lei Zhang
Chang'an University
Author Profile
Zhi Wang
Zhejiang University
Author Profile
Xiaonan Luo
Guilin University of Electronic Technology
Author Profile

Abstract

In the process of identifying non-line-of-sight (NLOS), acoustics-based indoor positioning needs to collect audio recordings of sound fields in multiple rooms and upload them to the central server for training. Once the transmission process and server-side suffer malicious attacks, private data will also be leaked. To solve the training difficulty and privacy issues at the same time, we propose a novel Personalized Federated Learning (PFL) model combined with user frequency and room data capacity, taking into account the significant differences in positioning data with room layout. The proposed model can accurately identify the differences between different room data when aggregating on the server-side. By collecting data in the actual indoor environment and comparing the existing algorithms, the accuracy of the proposed method in the data verification of unfamiliar rooms is 90%.
14 Feb 2023Submitted to Electronics Letters
15 Feb 2023Submission Checks Completed
15 Feb 2023Assigned to Editor
01 Mar 2023Reviewer(s) Assigned
11 Mar 2023Review(s) Completed, Editorial Evaluation Pending
13 Mar 2023Editorial Decision: Revise Minor
14 Mar 20231st Revision Received
15 Mar 2023Review(s) Completed, Editorial Evaluation Pending
15 Mar 2023Submission Checks Completed
15 Mar 2023Assigned to Editor
20 Mar 2023Editorial Decision: Revise Minor
25 Mar 20232nd Revision Received
27 Mar 2023Submission Checks Completed
27 Mar 2023Assigned to Editor
27 Mar 2023Review(s) Completed, Editorial Evaluation Pending
02 Apr 2023Editorial Decision: Accept