Abstract: In this paper, we first make use of the smoothing mean filtering method to smooth the RSSI value of the signal in order to improve the accuracy of signal acquisition.Secondly, we use k-means clustering algorithm to optimize the fingerprint database in order to improve the accuracy of the fingerprint database.Finally, we used a weighted approach to improve the traditional KNN algorithm and form a new fingerprint matching algorithm WKKN.The experimental results show that: Improved algorithm proposed in positioning accuracy and stability than traditional fingerprint Wifi positioning considered to have greatly improved.