Emitters Classification based on Steady State BT Signal Features

Abstract

Aims. The aims of this study were to verify the enhancement wireless networks’ security, and the identification of specific emitters. Methods. A novel technique was introduced to enhance the security of wireless network. The technique was built up based on the use of radio frequency (RF) fingerprinting for Bluetooth (BT) signals. Five emitters represented as mobile phones were considered to classify them. Two hundred and fifty BT signals were collected from these emitters. Portions of steady state of the BT signals studied. The applied methods were a detection method based on energy envelope was used to detect the portions, and the variational mode decomposition (VMD) method was applied to generate the intrinsic mode functions (IMFs). Features which represented as statistical information were extracted from the obtained IMFs. The Tree classifier was applied to the extracted feature to classify the introduced emitters. Results. The results demonstrated a high percent of correct classifications (93.5% through 100%). Conclusion. The results demonstrated the robustness of the features, the functionality, and the activeness of the introduced method; consequently, a high degree of wireless network security was achieved.