Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Improving WiFi CSI Fingerprinting with IQ Samples

Junjie Wang, Yong Huang, Feiyang Zhao, Wenjing Wang, Dalong Zhang, Wei Wang

TL;DR

CSI2Q tackles the challenge of performing RF fingerprinting with readily available CSI from commodity WiFi devices by transforming frequency-domain CSI into time-domain representations and transferring IQ-based feature extraction capabilities through auxiliary learning. The approach mitigates channel interference, compensates for coarse CSI granularity, and recovers fingerprint information lost in CSI processing. A dual-task model leverages an auxiliary IQ dataset alongside synthetic CSI data to improve device identification accuracy, achieving significant gains on both synthetic (up to 91%) and real (up to 82% average) CSI datasets. This work enables practical, hardware-light device authentication in wireless networks by bridging CSI-based fingerprints with IQ-based feature robustness. The results suggest CSI2Q can provide robust physical-layer authentication comparable to IQ-based methods while using readily available CSI from commercial WiFi gear.

Abstract

Identity authentication is crucial for ensuring the information security of wireless communication. Radio frequency (RF) fingerprinting techniques provide a prom-ising supplement to cryptography-based authentication approaches but rely on dedicated equipment to capture in-phase and quadrature (IQ) samples, hindering their wide adoption. Recent advances advocate easily obtainable channel state in-formation (CSI) by commercial WiFi devices for lightweight RF fingerprinting, but they mainly focus on eliminating channel interference and cannot address the challenges of coarse granularity and information loss of CSI measurements. To overcome these challenges, we propose CSI2Q, a novel CSI fingerprinting sys-tem that achieves comparable performance to IQ-based approaches. Instead of ex-tracting fingerprints directly from raw CSI measurements, CSI2Q first transforms them into time-domain signals that share the same feature space with IQ samples. Then, the distinct advantages of an IQ fingerprinting model in feature extraction are transferred to its CSI counterpart via an auxiliary training strategy. Finally, the trained CSI fingerprinting model is used to decide which device the sample under test comes from. We evaluate CSI2Q on both synthetic and real CSI datasets. On the synthetic dataset, our system can improve the recognition accuracy from 76% to 91%. On the real dataset, CSI2Q boosts the accuracy from 67% to 82%.

Improving WiFi CSI Fingerprinting with IQ Samples

TL;DR

CSI2Q tackles the challenge of performing RF fingerprinting with readily available CSI from commodity WiFi devices by transforming frequency-domain CSI into time-domain representations and transferring IQ-based feature extraction capabilities through auxiliary learning. The approach mitigates channel interference, compensates for coarse CSI granularity, and recovers fingerprint information lost in CSI processing. A dual-task model leverages an auxiliary IQ dataset alongside synthetic CSI data to improve device identification accuracy, achieving significant gains on both synthetic (up to 91%) and real (up to 82% average) CSI datasets. This work enables practical, hardware-light device authentication in wireless networks by bridging CSI-based fingerprints with IQ-based feature robustness. The results suggest CSI2Q can provide robust physical-layer authentication comparable to IQ-based methods while using readily available CSI from commercial WiFi gear.

Abstract

Identity authentication is crucial for ensuring the information security of wireless communication. Radio frequency (RF) fingerprinting techniques provide a prom-ising supplement to cryptography-based authentication approaches but rely on dedicated equipment to capture in-phase and quadrature (IQ) samples, hindering their wide adoption. Recent advances advocate easily obtainable channel state in-formation (CSI) by commercial WiFi devices for lightweight RF fingerprinting, but they mainly focus on eliminating channel interference and cannot address the challenges of coarse granularity and information loss of CSI measurements. To overcome these challenges, we propose CSI2Q, a novel CSI fingerprinting sys-tem that achieves comparable performance to IQ-based approaches. Instead of ex-tracting fingerprints directly from raw CSI measurements, CSI2Q first transforms them into time-domain signals that share the same feature space with IQ samples. Then, the distinct advantages of an IQ fingerprinting model in feature extraction are transferred to its CSI counterpart via an auxiliary training strategy. Finally, the trained CSI fingerprinting model is used to decide which device the sample under test comes from. We evaluate CSI2Q on both synthetic and real CSI datasets. On the synthetic dataset, our system can improve the recognition accuracy from 76% to 91%. On the real dataset, CSI2Q boosts the accuracy from 67% to 82%.
Paper Structure (21 sections, 18 equations, 5 figures, 1 table)

This paper contains 21 sections, 18 equations, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Threat Model and CSI Features
  3. Threat Model
  4. CSI Features
  5. System Design
  6. System Overview
  7. CSI Measurement Transformation
  8. Channel Interference Mitigation.
  9. Time-Domain Sample Generation.
  10. Feature Extraction and Recognition
  11. Experimental Evaluation
  12. Evaluation Methodology
  13. Auxiliary IQ Dataset.
  14. Synthetic CSI Dataset.
  15. Real CSI Dataset.
  16. ...and 6 more sections

Figures (5)

  • Figure 1: Workflow of CSI2Q. The dashed gray component exists only in the training phase.
  • Figure 2: Workflow of the auxiliary learning model.
  • Figure 3: Experimental setting for CSI collection in a meeting room.
  • Figure 4: Classification results of different models.
  • Figure 5: Confusion matrix of classification results on the real dataset.