Adaptive CEEMD-SVD Joint Denoising Algorithm via Aquila Optimization in FMCW Radar

Noise contamination in frequency modulated continuous wave (FMCW) radar signals significantly impairs target detection and results in inaccurate range and velocity measurements. To tackle this issue, an adaptive CEEMD - SVD joint denoising algorithm is put forward in this paper. By utilizing the aquila optimizer (AO) for noise reduction for the intermediate frequency (IF) signal of the FMCW radar, we get what is called as the AO-CEEMD-SVD denoising algorithm. Firstly, the original IF signal of the FMCW radar is decompose into a series of intrinsic mode functions (IMFs). Secondly, the Person correlation coefficient (PCC) method is utilized to identify IMFs demonstrating a significant relationship with the original signal. Thirdly, these selected IMFs are subjected to singular value decomposition (SVD), preserving a suitable count of the most significant singular values. The denoised signal is then reconstructed by integrating the processed IMFs with the residual components. Subsequently, the AO algorithm is utilized to construct a fitness function and search for the optimal parameter set for signal denoising. Finally, the optimal parameter set is used to perform CEEMD-SVD-based denoising and signal reconstruction to obtain the final denoised signal. Simulation experiments are conducted to validate the efficacy of the proposed AO-CEEMD-SVD denoising method, benchmarking it against current denoising techniques. The results indicate that the AO-CEEMD-SVD approach exhibits superior noise suppression performance and surpasses existing methods. Moreover, it achieves enhanced denoising performance for the FMCW radar signals.