FedPos: A Federated Transfer Learning Framework for CSI-Based Wi-Fi Indoor Positioning

Abstract

This paper proposes FedPos, a federated transfer learning framework together with a novel position estimation method for Wi-Fi indoor positioning. Compared with traditional machine learning with privacy leakage problems and the cloud model trained through federated learning (FL) fails in personalization, the FedPos framework aggregates non-classification layer parameters of models trained from different environments to build a robust and versatile encoder on the cloud server while preserving user privacy. The global cloud encoder can aggregate different classifiers and then construct personalized models for new users through fine-tuning. The proposed framework can be updated incrementally and is highly extensible. Specifically, we exploit channel state information (CSI) as the positioning feature and assess the transferability of a lightweight convolutional neural network (CNN) in unfamiliar environments. We evaluate the performance of our proposed framework and position estimation method in different indoor environments. Our experimental results indicate that the proposed framework can achieve a mean localization error of 42.18 cm in a 64-position living room. They also confirm that FedPos can achieve a 5.22% average localization performance boost and reduce the average model training time by about 34.78% when compared with normal training. By reusing part of the feature extractor layers that are trained from other environments, at least 65% of training data can be saved to achieve a localization performance that is similar to the base model. Overall, the proposed position estimation method can effectively improve localization accuracy as compared with three other existing CSI-based methods.

Multi-Level Split Federated Learning for Large-Scale AIoT System Based on Smart Cities

“…Federated learning is an emerging distributed machine learning paradigm that allows clients to jointly model without sharing data, breaking down data silos. With the deepening of research, systematic heterogeneity and statistical heterogeneity have become obstacles to the development of federated learning [16][17][18][19]. Karimireddy et al [20] proposed that the Scaffold algorithm corrects local model updates by adding variance reduction techniques to local training to approximately correct the drift of local training on the client side.…”

Hybrid Sensing for Fatigue Detection Using Wearables and RF

“…The received signal contains 64 subcarriers. The guard subcarriers are removed as they do not contain useful information [34]. Then the remaining subcarriers are divided by the sum of the absolute values of the subcarriers to remove the effects of Automatic Gain Control (AGC) which is present in the Raspberry Pi onboard Wi-Fi chip.…”

A Survey of Latest Wi-Fi Assisted Indoor Positioning on Different Principles

“…The neural network learned the relationship between signals and truth locations to build an environment model for positioning. Guo et al put forward a federated TL framework FedPos for CSI fingerprint-based indoor positioning scheme [ 51 ]. It aggregated the non-classification layer parameters of models trained from different environments to build a versatile encoder.…”

“…Similar work includes [ 107 , 108 ]. The latest work, the FedPos framework, also utilized the homomorphic encryption technology to address privacy issues [ 51 ]. However, the cryptography-based scheme incurs a large amount of computation and communication overheads.…”

“…Whereas it injected noise into the data, which might reduce the positioning accuracy. Recently, many works have introduced FL for privacy protection [ 51 , 110 , 111 , 112 , 113 ]. However, FL schemes suffer from the challenge of limited accuracy when applying the global model to different personalized scenarios.…”