Feasibility Study on Intra-Grid Location Estimation Using Power ENF Signals

Ravi Garg, Adi Hajj-Ahmad, Min WuarXiv 2021 · DOI · arXiv

Abstract

The Electric Network Frequency (ENF) is a signature of power distribution networks that can be captured by multimedia recordings. While previously used to identify broad regions, this study explores using ENF traces to pinpoint specific locations within a single power grid. By analyzing high-passed ENF signals, we demonstrate that correlation decreases as distance increases, enabling trilateration-based localization with a resolution of approximately 24 miles.

The Problem: Beyond Regional Identification

Forensic experts have long used ENF signals—subtle frequency fluctuations around 50/60 Hz—as a digital watermark to verify the time and general region of a recording. However, the prevailing assumption was that the ENF signal is uniform across an entire interconnected grid. This research challenges that notion, addressing the "intra-grid localization" problem: can we determine exactly where in the Eastern US a recording was made?

Methodology: Capturing Microscopic Traces

To isolate location-specific signatures, the researchers developed a specialized signal processing pipeline. First, the MUSIC (Multiple Signal Classification) algorithm extracts the instantaneous frequency with high precision. Then, a high-pass filter removes the global frequency trend, leaving behind "microscopic traces."

Zoomed view of ENF signals from three different cities, highlighting the microscopic variations that enable localization.

These traces are caused by local load changes and the finite propagation speed of frequency disturbances (roughly 500 miles/second). The team then applied two geometric approaches:

  1. Half-Plane Intersection: Comparing a query signal against anchor nodes to mathematically constrain the location to a region closer to the more highly correlated anchor.
  2. Correlation Quantization: A trilateration method mapping correlation ranges to distance "bins."

Key Insights: The Grid as a Positioning System

The core discovery is that the power grid has "latency." Because disturbances propagate at a finite speed, different geographical points experience the same event with slight temporal and intensity variations.

Correlation coefficients for different city pairs. Closer locations (College Park–Princeton) show significantly higher correlation than more distant pairs (Princeton–Atlanta).

In densely populated areas, the correlation of these high-frequency details serves as a reliable proxy for geographical distance. While the 60Hz "beat" is global, the "wiggles" on top are driven by neighborhood-level power consumption.

Results and Performance

  • Accuracy: In the Eastern US grid, the system achieved a localization resolution of ~24 miles using 15-minute segments.
  • Temporal Sensitivity: Longer recordings yield better results; an 8-minute segment significantly outperforms a 4-minute one.
  • SNR Requirements: For multimedia applications, a Signal-to-Noise Ratio (SNR) of at least 20 dB is required, with 30 dB matching the performance of direct power-line recordings.

Limitations and Future Outlook

The primary challenge remains the "Multimedia Gap." Forensic audio/video recordings often have an SNR of 9–15 dB, which is currently insufficient for fine-grained localization. Future efforts will focus on improving SNR extraction techniques, scaling the anchor node database, and exploring IoT applications for location authentication to prevent spoofing.

Summary updated

Abstract

The Electric Network Frequency (ENF) is a signature of power distribution networks that can be captured by multimedia recordings made in areas where there is electrical activity. This has led to an emergence of several forensic applications based on the use of the ENF signature. Examples of such applications include estimating or verifying the time-of-recording of a media signal and inferring the power grid associated with the location in which the media signal was recorded. In this paper, we carry out a feasibility study to examine the possibility of using embedded ENF traces to pinpoint the location-of-recording of a signal within a power grid. In this study, we demonstrate that it is possible to pinpoint the location-of-recording to a certain geographical resolution using power signal recordings containing strong ENF traces. To this purpose, a high-passed version of an ENF signal is extracted and it is demonstrated that the correlation between two such signals, extracted from recordings made in different geographical locations within the same grid, decreases as the distance between the recording locations increases. We harness this property of correlation in the ENF signals to propose trilateration based localization methods, which pinpoint the unknown location of a recording while using some known recording locations as anchor locations. We also discuss the challenges that need to be overcome in order to extend this work to using ENF traces in noisier audio/video recordings for such fine localization purposes.
BibTeX
@misc{garg2021feasibility,
  author = {Ravi Garg and Adi Hajj-Ahmad and Min Wu},
  title = {{Feasibility Study on Intra-Grid Location Estimation Using Power ENF Signals}},
  year = {2021},
  eprint = {2105.00668},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2105.00668},
}
Figure