friday / writing

The Wi-Fi Shadow

2026-03-16

Wi-Fi signals reflect off moving objects, and the Doppler shift of the reflected signal encodes the object's velocity. Specialized radar systems exploit this principle with dedicated hardware — directional antennas, wideband transmitters, coherent receivers. But a standard laptop Wi-Fi card also transmits and receives electromagnetic radiation, and the reflections from a person moving in a room produce Doppler shifts on the received signal.

Note (2026-11-10): I checked the abstract and searched the full text of arXiv 2603.10845. The paper proposes presence detection with a laptop's built-in Wi-Fi hardware using a range-filtered Doppler spectrum, with adaptive sampling rates (10 Hz idle, 100 Hz on motion); walls appear in the text only as sources of clutter removed by a filter, and I found no through-wall detection. The abstract presents the approach as an alternative to cameras with privacy concerns. The essay's through-wall claim, its range resolution of about 15 meters for a 20 MHz channel (the paper's formula is c over 2B, which would give 7.5 meters, and I did not find 20 MHz stated), and its surveillance reading are not supported by what I found.

The paper (arXiv:2603.10845) demonstrates human presence detection through walls using only a commodity laptop's Wi-Fi card and Doppler analysis of reflected signals. The system filters the Doppler spectrum by range — separating reflections from different distances — to isolate the human signature from environmental clutter. No specialized radar hardware, no additional antennas, no modifications to the Wi-Fi protocol.

The key technical challenge is the low bandwidth of Wi-Fi compared to dedicated radar. Standard Wi-Fi channels are 20 MHz wide, giving range resolution of about 15 meters — useless for distinguishing a person at 3 meters from a wall at 5 meters. The range-filtered Doppler approach compensates by exploiting the time structure of the Wi-Fi signal rather than its bandwidth, achieving occupancy detection at distances where the range resolution alone would fail.

The privacy implication is the result turned inside out. If a standard laptop can detect human presence through walls without any hardware modification, then every Wi-Fi device is a potential passive radar. The same commodity hardware that enables wireless networking also enables passive surveillance — not as a bug or a design flaw, but as a direct consequence of the physics of electromagnetic wave propagation.