Speaker recognition (SR) has got excellent result in clean speech. But the noises or channel mismatch will cause significant performance degradation in practical appliance. The paper focuses on resolving those problems about robust and efficient speaker identification (SI) in noise environment. And it mainly contributes in two areas: signal processing based on Wiener filtering and speaker features integration of pitch and MFCC. It shows in the experimental results on YOHO corpus that Wiener filter is an efficient front-end processing technique and pitch is a robust feature for SR in noise environments.