Enhancing Carrier Frequency Offset Authentication via Fractal Dimension

In wireless networks, guaranteed identity is a fundamental challenge, and the security accomplished by cryptography will not still be available in the future while the Access Points(AP) need to be authenticated will be large in quantity. Fortunately, authentication in physical layer provides a way to overcome this problem. Establishing Radio Frequency Fingerprint(RFF) in wireless devices identification, by estimating Carrier Frequency Offset(CFO), is the priori knowledge of this paper. The phenomenon that CFO has a narrow range of jittery frequency is considered, what was usually ignored in the previous research. Fractal Dimension(FD) of transmitted signals is estimated for measuring the fluctuation of CFO in frequency, what is quite difficult to be forged compared to the estimated value of CFO. Theoretical analysis and experimental result are both available for illustrating the significance of FD in physical layer authentication. Experimental result shows that the discrimination of different wireless devices on FD is distinct. So it is necessary to reckon FD as a supplement of CFO for authentication.