ØCLOAK隠

WiFi sensing infers activity from changes in radio signals. Research systems have demonstrated through-wall motion detection with inexpensive equipment. That interests me because the person being sensed may not own or control the equipment collecting the signal.

The capabilities vary with placement, antennas, frequency, and the surrounding room. Motion detection, breathing estimation, and pose reconstruction are different tasks. A result on one does not establish the others.

The project’s research notes cover WiFi sensing, Bluetooth trackers, device fingerprinting, and the limits of the proposed defenses.

The current design has two units. A mains-powered home unit would hold a switched reflector with enough physical area to affect radio paths in a room. A portable unit would carry the functions that need to travel with a person, including experiments with Bluetooth and WiFi decoys.

The reflector experiment asks whether changing those paths can reduce the accuracy of a sensing receiver while leaving ordinary communications usable. That benefit has to be measured at the intended frequency, size, and cost before it can become a product claim.

2.4 GHz

planned reflector test

efficacy remains unmeasured

4–32

elements in the sweep

compare several reflector sizes

The design excludes jamming. Reflection and decoy traffic still need technical testing and a review of the rules that apply to the eventual device.

A separate software idea is to generate coherent decoy activity that makes behavioural profiling less reliable. Random noise may be easy to filter, so the question is whether useful decoys can be generated more cheaply than an adversary can remove them.

A community network for sharing observations remains part of the longer-term plan. It would need a way to assess reports without exposing the people submitting them. Rotating identifiers or rounding a location would not, by themselves, establish anonymity.

The intended model is open hardware sold near manufacturing cost, with grants, crowdfunding, and donations supporting development. I don’t want a subscription or a VC-funded business that depends on collecting data from the people using it.

RF work can also have defense applications. The current plan puts consumer privacy first; counter-drone work is a possible later direction, not the funding source for this phase.

The repository contains research, product plans, bench protocols, and software tools. The hardware still depends on the reflector experiment. A successful result would be followed by prototype work, legal review, and certification before production.

An ultrasonic scanning tool is available for experiments with recorded audio. The browser illustration below shows the intended effect of obfuscation; it is not a measurement of ØCLOAK hardware.