TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Ask HN: Best technologies to track location of people in a large building?

3 pointsby eeemmmooo4 months ago
If I had a say 50,000 sq ft. Building that had up to 1000 people on it at one time. What is the best way to track where each of them are down to a couple inches accuracy. I’d also love to be able to get the direction they are facing as well. This tracking would of course be opted in and I assume would require some hardware that the person wears.

3 comments

hiAndrewQuinn4 months ago
A couple of <i>inches</i>? Realistically, cameras. Lots of cameras and some OpenCV to do the distance calculations.<p>Bluetooth triangulation is very hard (don&#x27;t ask me how I know) and won&#x27;t get you that kind of resolution. Apple&#x27;s AirTags won&#x27;t either. Dots painted on the floor at regular 1 yard or 10 foot intervals, in view of the cameras so that calculations can be inferred? Now we&#x27;re getting somewhere.<p>The good news is security cameras have existed for a long time, and you can almost certainly find someone to bulk purchase from on the cheap.<p>If you insist upon using a wearable you may be able to get a light source on everyone you can calibrate towards. I&#x27;m not sure whether that&#x27;s very helpful in practice.
评论 #42751399 未加载
runjake4 months ago
In the real world, you aren&#x27;t going to track them down to a couple inches accuracy. There&#x27;s too much variance in physical materials and positioning, let alone the RF environment.<p>With that caveat, to answer your question, you&#x27;d probably use Bluetooth LE devices that everyone carries. If everyone had wi-fi devices, you could use wi-fi triangulation software. This is usually an add-on package to existing enterprise wireless solution.<p>You could also do active NFC tags and have choke-points with scanners everywhere.<p>I have my own Python script at work, appropriately named creep.py, that locates people by wi-fi by logging into our enterprise wireless system, checking what AP their MAC&#x2F;DHCP client ID&#x2F;username combo is connected to, gets the RSSI, then runs that through some if&#x2F;else statements to determine what room they&#x27;re in to a high degree of accuracy. The script also gets readings off the adjacent APs, as well, for better triangulation. I keep quiet about this script, lest it be used for police state purposes.<p>The above is a simplification of the script, but if RSSI is approx &lt;=55 or so, then the person is likely in the same room as the WAP, given normal US construction practices. ~65 or so, there&#x27;s probably a drywall wall between the person and WAP, etc etc. This is an extreme oversimplification, but gives you an idea.<p>tl;dr:<p>Your tracking options are:<p>- Bluetooth LE<p>- Wi-Fi<p>- NFC<p>- GPS devices<p>None of these will get you to a couple inches of accuracy. Think more like &quot;room-level accuracy&quot;.
markus_zhang4 months ago
What is the business case?
评论 #42745162 未加载
评论 #42751340 未加载