Under specific driving conditions, the airflow passing through the vehicle\'s side mirrors can generate high-frequency noise with narrow-band frequency characteristics, commonly referred to as whistling. The causes of whistling are multifaceted and include factors such as gaps, shape, and periodic airflow motion. Different mechanisms of whistling generation necessitate distinct solutions. In this paper, whistling was detected during subjective evaluation in the wind tunnel. Through wind tunnel testing and acoustic array analysis, the frequency of the whistling was identified, and the airflow velocity and yaw angle were determined. Subsequently, numerical analysis of the wall-bounded airflow near the side mirrors was conducted using a transitional transport function incorporating intermittency and momentum thickness Reynolds number corrections. The analysis revealed that the separation and reattachment of the laminar boundary layer led to localized turbulence enhancement, which was identified as the cause of whistling. Based on these findings, two optimization schemes for promoting early transition of the laminar flow were proposed and validated through simulations. The simulation results demonstrated that designs incorporating steps or modifications to surface roughness could effectively prevent the separation of the laminar boundary layer, thereby eliminating the generation of whistling. Moreover, the simulation results were found to be consistent with the experimental findings. This study not only elucidated the mechanism of whistling generation due to the shape of the side mirrors but also provided a theoretical basis and technical support for the design optimization of vehicle side mirrors.