Estimating CPU Features by Browser Fingerprinting

Browser fingerprinting is getting popular in cutting edge Web developers. It typically uses the header information, such as user-agent and plugins. However, the header information can easily be modified or altered by configuration or a browser's extensions. Unlike the header information, hardware information is difficult to be changed, and is regarded as valuable information in browser fingerprinting. Therefore, in this paper, we propose a method that infers the presence or absence of a CPU extension called Advanced Encryption Standard New Instructions (AES-NI) and Intel Turbo Boost Technology (Turbo Boost). Theoretic analysis and experimental results indicate that this method is efficient and feasible for browser fingerprinting.