Sim2Real Deep Transfer for Per-Device CFO Calibration

TL;DR

This paper introduces a transfer learning approach that combines simulation pretraining with minimal real data fine-tuning to improve CFO estimation across different SDR devices, significantly reducing error rates.

Contribution

It presents a novel Sim2Real transfer learning framework for CFO calibration that adapts a pre-trained DNN to specific hardware impairments with minimal real data.

Findings

Achieves 30x BER reduction compared to traditional methods.

Effectively adapts to multiple SDR platforms with minimal real data.

Bridges the simulation-to-reality gap for robust CFO estimation.

Abstract

Carrier Frequency Offset (CFO) estimation in Orthogonal Frequency Division Multiplexing (OFDM) systems faces significant performance degradation across heterogeneous software-defined radio (SDR) platforms due to uncalibrated hardware impairments. Existing deep neural network (DNN)-based approaches lack device-level adaptation, limiting their practical deployment. This paper proposes a Sim2Real transfer learning framework for per-device CFO calibration, combining simulation-driven pretraining with lightweight receiver adaptation. A backbone DNN is pre-trained on synthetic OFDM signals incorporating parametric hardware distortions (e.g., phase noise, IQ imbalance), enabling generalized feature learning without costly cross-device data collection. Subsequently, only the regression layers are fine-tuned using $1,000$ real frames per target device, preserving hardware-agnostic knowledge while adapting to device-specific impairments. Experiments across three SDR families (USRP B210, USRP N210, HackRF One) achieve $30\times$ BER reduction compared to conventional CP-based methods under indoor multipath conditions. The framework bridges the simulation-to-reality gap for robust CFO estimation, enabling cost-effective deployment in heterogeneous wireless systems.