Active Surface Electromyography Adaptive Filtering Algorithm Based on Sparrow Search*

Attempting to solve the problem that active surface electromyogram (sEMG) signals contain a large amount of random noise, resulting in poor quality of the collected signals, this paper proposes an adaptive noise canceling (ANC) filter based on the Sparrow Search Algorithm (SSA) to remove mixed random noise in active sEMG signals. The SSA-ANC filter employs the SSA to optimize the weights of the adaptive filter, enhancing the filtering effect on sEMG signals. The noise reduction performance of the SSA-ANC filter proposed in this paper is evaluated through standard metrics such as Signal-toNoise Ratio (SNR), Mean Squared Error (MSE), and Maximum Error (ME). The experimental results show that the average SNR of the active sEMG signals processed by the proposed SSA-ANC filter can reach $\mathbf{7 5. 6 4} \mathbf{d B}$ when the input SNR is $\mathbf{1 0}$ dB. Compared with other adaptive filters based on evolutionary algorithms, this represents an 11.07% improvement, indicating a significant performance enhancement.