Improved Artificial Rabbit Optimization and Its Application in Multichannel Signal Denoising

The application of multichannel signals, along with the development of sensors, has greatly improved the efficiency of modern communication and data processing systems. However, noise interference between different sensor channels often leads to poor signal quality in multichannel signals, impacting the reliability of communication systems. Multichannel signal denoising is beneficial for improving signal quality, making signal analysis and recognition in complex environments more feasible and effective. Most of the previous noise reduction methods focus on single-sensor signals, and it is difficult to deal with the fusion noise of multisensor signals. This article proposes a multichannel signal denoising method based on improved artificial rabbit optimization (GCARO) and successive multivariate variational mode decomposition (SMVMD) algorithm, called GCARO-SMVMD. In this article, an improved GCARO based on the golden sine and Cauchy mutation is employed to optimize the penalty factor of SMVMD. The optimized SMVMD is then utilized to decompose the multichannel signal, effectively segregating the noisy mode from the noiseless mode, and reconstructing the signal using the noiseless mode. Simulation results demonstrate that the signal-to-noise ratio of the proposed method has improved by at least 14 dB. In real-world signal experiments, the proposed method exhibits the most effective denoising performance, as evidenced by the clear and smooth trajectory of the attractor. Additionally, detrended fluctuation analysis (DFA) results indicate high signal quality.