Feature Extraction Based on Frame Interval for Wireless Network Devices

Abstract

The development of wireless communication technology has brought great convenience to our lives, however, in the fields of military communications, remote signal control and wireless signal transmissions, we still face huge challenges. Based on the problems above, some researchers extract features from the time domain and the frequency domain of the transient and steady-state part of the wireless signal separately, ultimately achieved the purpose of identification individual wireless network devices; some researchers extract the features of the wireless frames by parsing the IEEE802.11 protocols, and the method can also achieve the purpose of identifying wireless network devices. For the steady-state part of the wireless signal, it needs high-precision equipments for data acquisition, and the volume of data obtained is very large. As for the transient part of the wireless signal, it has a very short duration, conventional equipment can hardly meet the requirements. The method by parsing the wireless frames and then extracting the frame interval is also very inefficient, it has some limitations for parameter acquisition. In this paper, we proposed a method which takes frame interval as a fingerprint to represent wireless device, this method eliminates the need for high-precision equipments, at the same time, it avoids the demands to parse the IEEE802.11 protocols. Using this method, we can quickly and easily get data whose volume is quite small without expensive equipment. Probability density curves are used in this paper to represent the signature. The experimental results show that the proposed method is effective for the identification of IEEE802.11 wireless network devices, and the average recognition rate reaches 95%.