Multiple Large networks - project possibility evaluation

Hello,

I’d like to verify my project idea before starting the implementation.

So far, I do have the MDEK Dev Kit. Tests showed, that in the environment in focus (industrial building) an anchor every 7.5m in x and y delivers acceptable to good localization results.

The questions are:

  • What module (e.g. dwm1001) shall I choose if the areas in interest are large, e.g. 60m x 130m which would result in 144 anchors in the given distance if all anchors were in one network (preferred solution). The project total would be 5 areas, each approx. 8.000m² or 40.000m² combined with up to 1.000 active tags, communicating over an ESP device via MQTT to AWS Cloud.

  • If one network per area is not possible due to limitations in the number of anchors per network, what is another way of achieving an UWB (Decawave) based RTLS solution for those areas, e.g. multiple networks.

  • If the solution is multiple networks per area, what preconditions have to be fulfilled? I read in other posts, that in this case none of the anchors from network A shall see any anchors of network B, but I am not sure whether I got that right.

  • Last question: Given multiple networks, either per area or (preferred) per building. If a tag moves out of network A into network B. Will it be automatically recognized by that network (e.g. tagged object moved from building A to building B) and what is the maximum number of tags given a network of either 144 anchors or the maximum number of anchors allowed per network.

Thank you,

Chris

Hi @ChristianFFM
7.5m x 7.5m grid seems to be a bit small to me, but it depends on environment.

Regarding your questions:

  1. From your comment I assuming that you are using PANS for evaluating you area. Only the DWM1001 module is supported by PANS binary.

  2. There is no limit on anchors count per network. But the anchors have some limit regarding anchors in range, each anchor should not hear more than 29 other anchors (regardless on multiple networks). Typically this rule should not be a problem to fulfill, but it depends on environment. See the document
    DWM1001_System_Overview.pdf - chapter 5 Scalability

  3. As noted above, there is a surrounding anchor limit, increasing the number of networks will not help here.

  4. Each tag is assigned to one network. So transition tag from network A to network B will cause that the tag will not work in network B.

Regarding tag number) The system capacity is 150Hz in one cluster. (you can have 15 tags @ 10Hz, or 750tags at 0.2Hz) For better system understanding see the whole document DWM1001_System_Overview.pdf .

Regarding AWS) With the regular PANS MQTT you need to be careful abut the amount of messages sent to AWS server as the system it self can generate very large amount of messages (150Hz * 60 * 60 = 540k MQTT messages per hour).

Cheers
JK