This paper addresses the first investigation and analysis of electronic network frequency (ENF) signals in the Tokyo metropolitan area, Japan. Electric power signals are recorded directly from a clean power line at seven different sites simultaneously for several weeks. Instantaneous frequency measurements based on time-domain analytic signals are performed on bandpass-filtered electric power signals, therein providing higher temporal resolution compared with the conventional FFT-based method combined with quadratic interpolation for extracting ENFs. Spectro-temporal analysis of the fluctuations of the ENF signals reveals that temporal correlations between the fluctuation energy in the frequency range of 0.4 Hz to 1.0 Hz obtained at different sites are inversely correlated to the geographical distances between the sites. The similarities of the spectro-temporal ENFs obtained from different sites show generally higher correlations with the geographical distances than the similarities of high-pass-filtered ENFs. Location estimation using linear regression between the similarities of spectro-temporal ENFs and the geographical distances of the anchor sites predicts the location of a target site with a mean prediction error of approximately 20 to 30 km.
https://www.aes.org/e-lib/browse.cfm?elib=18440
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.
Learn more about the AES E-Library
Start a discussion about this paper!
