An Simulated Annealing Artificial Rabbit Optimization Algorithm and Its Application in Weak Feature Extraction of Harmonic Radar

Nonlinear target detection in strong noise backgrounds requires higher sensitivity for harmonic radar. To address the issue of low signal-to-noise ratio (SNR) in echo signals and poor stochastic resonance parameters, this paper proposes an adaptive stochastic resonance system based on the Simulated Annealing Artificial Rabbit Optimization (SAARO) algorithm and an underdamped tri-stable stochastic resonance (TSR) model. First, an adaptive stochastic resonance system is constructed using the artificial rabbit optimization algorithm and the underdamped tri-stable model. The Fitness Distance Balance (FDB) strategy and the Metropolis criterion are then incorporated to enhance the algorithm's ability to escape local optima. Experimental results on six benchmark functions show that SAARO outperforms other traditional methods. Further tests on simulated and actual radar echo signals demonstrate that the SAARO-TSR method enhances weak signals in strong noise, improving the SNR by at least 12 dB by optimizing system parameters.