Interference Analysis and Successive Interference Cancellation for Multistatic OFDM-based ISAC Systems

TL;DR

This paper analyzes mutual interference in multistatic ISAC systems sharing spectral resources and proposes a low-complexity successive interference cancellation method that enhances sensing and communication performance.

Contribution

It introduces a novel adaptive interference cancellation technique for multistatic ISAC systems, improving BER, EVM, and radar image quality with low computational complexity.

Findings

Interference impacts BER, EVM, and radar imaging.

The proposed cancellation method reduces interference effectively.

Experimental results confirm performance improvements in practical setups.

Abstract

Multistatic integrated sensing and communications (ISAC) systems, which use distributed transmitters and receivers, offer enhanced spatial coverage and sensing accuracy compared to stand-alone ISAC configurations. However, these systems face challenges due to interference between co-existing ISAC nodes, especially during simultaneous operation. In this paper, we analyze the impact of this mutual interference arising from the co-existence in a multistatic ISAC scenario, where a mono- and a bistatic ISAC system share the same spectral resources. We first classify differenct types of interference in the power domain. Then, we discuss how the interference can affect both sensing and communications in terms of bit error rate (BER), error vector magnitude (EVM), and radar image under varied transmit power and RCS configurations through simulations. Along with interfernce analysis, we propose a low-complexity successive interference cancellation method that adaptively cancels either the monostatic reflection or the bistatic line-of-sight signal based on a monostatic radar image signal-to-interference-plus-noise ratio (SINR). The proposed framework is evaluated with both simulations and proof-of-concept measurements using an ISAC testbed with a radar echo generator for object emulation. The results have shown that the proposed method reduces BER and improves EVM as well as radar image SINR across a wide range of SINR conditions. These results demonstrate that accurate component-wise cancellation can be achieved with low computational overhead, making the method suitable for practical applications.