Hi,
We have developed a solution using MDEK1001.
I am looking for commercial options for TWR based anchors/tags using DWM1001 that can offer longer range, higher tag capacity and better accuracy. TWR based positioning support is a must.
Can someone please suggest available options in the market?
Regards,
Ram
TWR and high tag capacity don’t combine well and implies a low update rate.
TWR requires more radio transmissions than TDoA which is why most high tag count systems use TDoA. TDoA requires more complexity in the anchors (they all need to be time synchronised for a start) but simplifies the tags. If you have lots of tags then that’s exactly what you want.
I can point you to a commercial TWR based system that can cover large areas (currently largest install is over 28,000 square meters) and can give accuracies of 2-3 cm. But it only supports a small number of tags, it’s designed for a small number of items at a high update rate and low latency. So probably not what you want.
Also if you are looking for a commercial system why do you care about the positioning method used? A commercial system will be locked down, it’s unlikely they will give you access to the underlying raw data. If all you get is a position why does it matter how it was generated?
Thanks Andy for your detailed response.
We need TWR because the tags need to know their position in our application without the help of any centralized server to provide coordinates.
Can you please point me to the commercial TWR solution that you mentioned?
It’s sold under two different names depending on the application, tracking cars or cameras, links below.
The core technology is the same for both, 100Hz position output on CAN and serial with a fixed <60ms latency (you can reduce that but the output gets nosier). If you get the setup right you can get a mean 2D position error in the 2-3 cm region. Height accuracy very much depends on the geometry.
Optional built in inertial integration which reduces the noise on the outputs, especially the speed and height outputs.
The big down side is that it is only 5 tags (or rovers as they are called) at a time. That was an intentional design compromise, we needed to be able to accurately track things pulling a couple of g. When you’re trying to do that there is a limit to how many device you can share time between.
Also it’s not the cheapest system, they’ve got to pay my wages somehow 
Hopefully obvious disclaimer - I designed a lot of the system and have an indirect financial interest in how many are sold.
There are a few other systems that I think are similar - Pozyx, Humatics, Kinexon but I don’t think any of them put such an emphasis on both dynamics and low/constant latency outputs.
For automotive use:
For camera tracking:
Thanks Andy. Very interesting. That’s a super customized solution 
We indeed need at least a decent capacity offered by the basic RTLS TWR solution (i.e. 15 tags at 100msec sampling).
Among range, capacity and accuracy, if someone offers better range than the MDEK1001 solution, rest being the same, that would interest us as well. Of course, considering large scale deployments, cost consideration would always be there.
Unfortunately in order to get better range you’re probably looking at a different antenna design. As soon as you change the antenna you get into all sorts of regulatory issues if you want to sell it commercially.
Commercial products tend to fall in to two groups, cheap ones that are a DWM1001 running PANS in a box with a thin software wrapper added. And expensive ones that use different hardware/firmware where they have done the required testing and certification.