Browsing by Author "AL-MASHAQBEH, HEMAM"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
Item USE OF WAVELET DECOMPOSITION IN RADIO FREQUENCY FINGERPRINTING OF BLUETOOTH SIGNALS(2022-01-10) AL-MASHAQBEH, HEMAM; Dalveren, Yaser; Kara, AliThis thesis addresses a new Open Systems Interconnection (OSI) Physical (PHY) layer scheme for extract and exploits Radio Frequency Fingerprinting (RFF) to uniquely identify specific devices such as cell phones based on their Bluetooth (BT) signals. Firstly, fingerprint features were extracted from BT transient signals experimentally collected from cell phones. After analysing transient Bluetooth signals, Dual-Tree Complex Wavelet Transform (DT-CWT) has been used to decompose Bluetooth signals. Feature extraction was performed from both time domain (TD) and wavelet-domain (WD) signals. Then, for classification, supported vector machine (SVM) classifier was used. Next, classification results achieved for time domain (TD) and wavelet domain (WD) BT signals were compared. The experiments were performed under different transient durations with different SNR levels such as low SNR (0 < SNR< 5 dB), moderate SNR (5 < SNR < 15 dB), high SNR (15 < SNR < 25 dB), and very high SNR (25 < SNR < 35 dB). Results show that it is possible to achieve reasonable accuracy in WD (at least 88%) even with short transient durations at low SNR levels. When compared to the results achieved for TD BT signals, better detection accuracy is clearly observed for WD BT signals. Therefore, it concluded that the use of DT-CWT can be evidently used in RFF of BT signals.