Indoor Location Detection using Wifi | Marko Tisler | WLPC EU Budapest 2016
Вставка
- Опубліковано 12 лип 2024
- Chapters:
00:00 - Start
01:17 - Expectation vs. Reality: Accuracy
03:07 - Expectation vs. Reality: Time
04:54 - Customer Expectations vs. Reality
06:36 - RSS Based Methods
08:59 - RSS - Theory vs. Reality
10:48 - RF Fingerprinting
13:25 - Post- Processing
17:50 - RSS - Strengths and Weaknesses
18:46 - Time Based Methods
19:38 - Time Based Methods - Road Trip Time (RTT)
20:21 - Cumulative Distribution Factor
22:06 - Time Based Methods - Time Difference of Arrival (TDoA)
23:22 - Time Based Methods - Synchronous
23:48 - Time Based Methods - Asynchronous
24:54 - Time Based Methods - Strengths and Weaknesses
25:58 - Angle of Arrival (AoA)
27:46 - AoA - Strengths and Weaknesses
28:24 - Hybrid Methods
30:25 - Exposing that Data
33:23 - Question and Answer
This is one of the best RTLS presentations I watched. Pretty factual.
Thanks for watching, Draxy!
It's amazing. I have learned basic knowledge of Indoor Location easily. Really impressive.
Thanks a lot, Muhammad! 😊
Only by switching to 5GHz I guess you can get more reliable position since its dispersion of signal across some floor plan should be more predictable because it does not really penetrate walls and objects like 2.4GHz does. Of course it makes it harder when you consider that only by closing door or by standing in front of an antenna can alter signal so much but as a guy explained moving target gives a lot more info which can be filtered and converged into a usable location data.
Also, since more and more APs are utilizing beam forming by only using a couple pf antennas I guess by creating an AP with a precisely manufactured phased array antenna you should be able to get much better client location accuracy.
For those who dont know what beam forming and phased array antennas are, I highly recommend looking it up on youtube, its supper interesting.
Reckon Point is getting meter accuracy using wifi and magnetic data, we can tell direction whether you're looking at the milk or coffee. We use the sensors in smartphones and compare the data of the building that we have mapped out with our proprietary robot. Would love to chat if anyone is interested,
Add in some pedestrian dead reckoning, Kalman filters, and other algorithms to get greater accuracy. We are shooting for 30 cm accuracy indoors with no additional infrastructure.
J.C. Milam hellooo can you tell me more details about this?
very informative
Glad it was helpful. Thanks for watching, Mohammad!
Now in 2018 Google will be implementing IEEE 802.11mc for indoor positioning directly in Android as a framework any developer can tap into, so it won't be app-specific.
If u control the client's and APs u can put them in low power mode for smaller spaces for a more accurate approximation? Am I right?