Transient fingerprint feature extraction for WLAN cards based on polynomial fitting

With the rapid development of cognitive radio network (CRN), the information security has become a primary concern. In this paper, transient fingerprinting is incorporated into CRN for detecting Primary User Emulation (PUE) attacks. A novel approach is proposed to extract transient fingerprint feature of transmitters. The energy envelope obtained from the instantaneous signal by spectrogram analysis, is fitted according to the least square estimation. The polynomial coefficients are selected as the feature vectors of instantaneous signal. A data acquisition system is designed to capture IEEE802.11b signals in the 2.4GHz ISM band. The results of data analysis indicate that it greatly improves the performance of detecting different brand of WLAN cards.