Autoregressive Model-based Signal Reconstruction for OCDM-induced Interference Mitigation in FMCW-based Sensing Systems

In this paper, we propose an autoregressive (AR) model-based signal reconstruction method to mitigate interference generated by orthogonal chirp division multiplexing (OCDM) signals in frequency-modulated continuous wave (FMCW)-based sensing systems. Recently, the emergence of OCDM as a promising integrated sensing and communication waveform has introduced coexistence scenarios with conventional FMCW-based sensing systems. Such coexistence results in significant mutual interference between the two systems, which degrades target detection performance. To address this issue, we adopt a power-based thresholding method that removes interference-contaminated regions in the fast-time FFT domain. Subsequently, the FMCW signals removed by the thresholding process are reconstructed using an AR model-based signal reconstruction method. As a result, the reduction in the mean false alarm rate achieved by the proposed method is 85.38% greater than that achieved by the B-spline-based method over the evaluated signal-to-interference ratio range. This result demonstrates the effectiveness of the proposed method in mitigating OCDM-induced interference in FMCW-based sensing systems.