Hacker News new | past | comments | ask | show | jobs | submit login
Passive Wi-Fi: Bringing Low Power to Wi-Fi Transmissions (usenix.org)
136 points by godelmachine on May 12, 2018 | hide | past | web | favorite | 29 comments



There's been a few links regarding WiFi backscatter use around here recently.

Connecting RF-Powered Devices to the Internet : http://iotwifi.cs.washington.edu/

3-D printed objects connect to WiFi without electronics : http://www.washington.edu/news/2017/12/05/in-first-3-d-print...


I also remember an article about a very low power camera performing read operations in an analog way with no onboard power. I think. I can't find the article though.

Also there was https://github.com/seemoo-lab/mobisys2018_nexmon_software_de....

If you think about it, there are a lot more things that can be done with wifi if you can arbitrarily TX/RX signals. In fact I'm getting a bit scared now.


You may be thinking of this article: https://news.ycombinator.com/item?id=16881946


Yes, this is it.


Camera one was a borderline scam (used middle out compression), you have nothing to be scared of if all of UW research is of similar "quality".


Can you justify that? It doesn't seem like that was the consensus on the HN thread about it.


Read the actual research paper?


The paper isn't self-evidently fradulent to me. Since it wasn't part of the discussion in the dedicated hn thread it seems like you just have some special insight to be able to make such a strong statement?


Oh, so claiming "60 fps 720p and 1080p" and later mentioning you are actually transferring ~100x50 and throwing out the rest is ok with you. Read that paper again.


Not sure where you're getting that "~100x50" figure, do you literally mean they're transferring a 100-pixel-wide image 50 times per second?

The closest thing I can see them mention in that paper is their adoption of the "zig-zag scanning" rather than raster scanning. But how many pixels that skips over is very variable dependent on the input image.

There's also the "super-pixels" thing, but that's pretty standard for video compression right? Macro-blocks and B/P-frames, corrected with occasional full-resolution I-frames...

They also note the effect of distance, both in the abstract and in the results.. 720@60 at 10 feet, but only 480@10 at 16 feet. Not remarkable results, but they're not lying in their abstract...


Supposed 1080 video would require ~100x50 resolution image (20x20 super-pixels) transmitted 30 times per second, rest of the ~3Mbit of bandwidth would be dedicated to actual video data. 3Mbit is enough for one "I" frame every 5 seconds, paper speaks of sending I frames every 8 seconds. That leaves 140KB per second for your magic compression, incidentally LESS than required by 100x50@30 L frames alone, and what about actual useful data? We didnt even mention color (~3x the data).

Lets try 720@10fps. Budget is 3MB per 8 seconds. 64x36@10 L frames = 184KB, 921KB I frame, leaves us with 236KB/s, enough BW to transfer at max only ~20% of changed super-pixels per second. Even their simulated demo video doesnt appear to be real.

Things just dont add up.


This seems to require a near-by powered wifi device to aide in association of the passive device. For large deployments, that would require many powered units. In addition to that, there are requirements for multiple MACs, which may also lead to problems with limits in some networks. This is all clever, but it doesn't seem practical to me since the passive device depends on a near by powered device to do anything. I have worked on sensor networks with battery powered wifi devices. It's certainly tricky.


>This seems to require a near-by powered wifi device to aide in association of the passive device.

...Yes? Gateways (routers) are there anyways.

>For large deployments, that would require many powered units.

As in all deployments?

>In addition to that, there are requirements for multiple MACs, which may also lead to problems with limits in some networks.

Wot? This is a non issue.

>This is all clever, but it doesn't seem practical to me since the passive device depends on a near by powered device to do anything. I have worked on sensor networks with battery powered wifi devices. It's certainly tricky.

Then you don't understand what this is about.

Backscatter is awesome because you literally don't put energy into generating rf yourself.


>View the slides

Can't we have PDF.js decode and render .pptx files? I bet they're no more complex than PDF.


You're betting wrong. The office formats are horrifically complex.


Is this system not susceptible to over-the-air attacks by malicious adversaries?


If I'm understanding it correctly then the answer is not much more than our present system. You get low power consumption in your device by being parasitic on a high-power monochromatic antenna furnished by the local WiFi.

There might be interesting attacks with other high-power antennas in the region creating pockets of destructive interference -- essentially localized denial-of-service type stuff.

The other risk you're taking is something like "the operator of this antenna can induce bit flips and such in your communications," which is true, but they wouldn't know what bits they're meddling with because they have to commit to the inconsistent signal before you use it. But they can already send arbitrary local packets pretending to be you at that granularity. Certainly if we're talking about the signal source managed by the WiFi operator, there is no dispute that this doesn't empower them -- they own the router, where they already have direct control over your internet communication and what they "hear from" and "say to" you.

Compared to the threat of opening your own access points with a suggestive name and just letting people connect to you, which you can also do with this range, being able to meddle with packets in a haphazard way doesn't seem very scary.


And why's it always malicious adversaries? Why can't we ever get some benevolent adversaries around here?


People want to defend against the latest threat.

Now, what on earth is a benevolent adversary?


I thought of something else that is not discussed often. There is a class of actors that can do good actions with bad ulterior motives.

For example in France we had a group of anti-muslim people provide free food to the homeless but the food had specifically pork in it.

Or linked to a recent HN story a vegan terrorist could infect himself with that tick that renders you allergic to meat and then go donate blood for saving children.


Those are really interesting examples although obviously stretching the definition of "benevolent."


Do you have links to either of these stories?


The Drill Instructor who, as you become more capable, increases the intensity of the challenge, is a benevolent adversary.


> Now, what on earth is a benevolent adversary?

If you find yourself a target of one, it means you're most likely evil.

(Also, in less extreme sense, I'd classify hackers as such, i.e. people exploiting systems for fun and learning, without doing any harm to anyone.)


Penetration Tester would seem to typify benevolent adversary.

Or anyone who's role it is to test the sturdiness of a system or product etc.


Someone that hacks your bank account to put more money in it?


Parents and other teachers who simulate and stage adversity to develop student skills.


Like the recent Internet Chemotherapy?





Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: