In this article, the authors investigate wide-area frequency as a criterion for digital audio recording authentication.
This paper briefly introduces the Frequency Monitoring Network (FNET) at UTK and analyzes the frequency characteristics of the four North American interconnections, concluding that the Short-time Fourier transform (STFT) can be used as an accurate Electrical Network Frequency (ENF) extraction method. The ENF criterion is a novel method for digital audio recording authentication in the field of forensic science. Both an accurate frequency estimation method and a reliable frequency reference database are the key requirements for this technique. Wide-area frequency measurements in each interconnection conform to the Gaussian distribution, but with slightly varied parameters. STFT is adopted to estimate the power system frequency signal embedded in audio files, and a procedure for using the ENF criterion, ranging from signal preprocessing to frequency estimation and frequency data matching, is proposed and then tested by two cases. Furthermore, factors which influence the accuracy of frequency estimation, such as the signal-to-noise ratio (SNR) and the recording hardware, are also discussed. (Published Abstract Provided)