Experimental Study on Reference-Path-Aided System Calibration for mmWave Bistatic ISAC Systems

TL;DR

This paper proposes a reference-path-aided calibration scheme for mmWave bistatic ISAC systems, effectively compensating system errors and enabling precise sensing even with LoS blockage.

Contribution

A novel calibration method leveraging reference path identification and error compensation to improve sensing accuracy in bistatic ISAC systems.

Findings

Supports Doppler and delay estimation with errors within 1 nanosecond

Utilizes delay-angle sparsity for reference path identification

Enhances synchronization robustness in mmWave ISAC systems

Abstract

Integrated sensing and communications (ISAC) has been regarded as a key enabling technology for next-generation wireless networks. Compared to monostatic ISAC, bistatic ISAC can eliminate the critical challenge of self-interference cancellation and is well compatible with the existing network infrastructures. However, the synchronization between the transmitter and the sensing receiver becomes a crucial problem. The extracted channel state information (CSI) for sensing under communication synchronization contains different types of system errors, such as the sampling time offset (STO), carrier frequency offset (CFO), and random phase shift, which can severely degrade sensing performance or even render sensing infeasible. To address this problem, a reference-path-aided system calibration scheme is designed for mmWave bistatic ISAC systems, where the line-of-sight (LoS) path can be blocked. By exploiting the delay-angle sparsity feature in mmWave ISAC systems, the reference path, which can be either a LoS or a non-LoS (NLoS) path, is first identified. By leveraging the fact that all the paths suffer the same system errors, the channel parameter extracted from the reference path is utilized to compensate for the system errors in all other paths. A mmWave ISAC system is developed to validate our design. Experimental results demonstrate that the proposed scheme can support precise estimation of Doppler shift and delay, maintaining time-synchronization errors within 1 nanosecond.