Best way to track tag outside of anchor station

Wireless Connectivity Ultra-Wideband
, ,
1

Hi All, Im looking for some advice for a unique situation, a little different to the regular tracking within a room full of anchors.

In short I want to create a base station that will form my anchor. The tag would move around this anchor and I would be able to track both distance and position of the tag. The tag would operate within 5-10m of the anchor and accuracy needs to be accurate to around 0.25 - 0.5m. Both the tag and base station have a BT BLE link for other functions.

The base station ideally would be about 300mm diameter and include x1 UWB module with x3 Antenna. (trying to keep cost to a minimum and avoid the use of multiple modules).

My question apart from the feasibility of the above is what is a the best method to use for tracking and what type of hardware is recommended for the base station.

For efficiency i would like to use PDoA so the tag is sending blinks at regular intervals. These are received by the x3 antenna presumably at different times and location is determined.
I am struggling to understand how the distance is then determined?
It is not feasible to add a clock to each device to synchronise timestamps but i could start a clock on the anchor & tag at the same time (initiated by the BLE from the tag) assuming there would be some error between the clocks (but this would be minimal over the operating distance?).

I have looked at TOF and TDoA but assume these will require more than x1 module in the base station?

I may be overcomplicating this but its been hard to find any examples of tracking a device like this.

As always, appreciate any help/explanation.

2

The PDoA version of the DW3000 only supports 2 antennas. So I’ll assume that for the initial answers:

Getting 25-50 cm range accuracy over 5-10 m is not an issue, getting range accuracies in the 10 cm region is relatively straight forward, it’s when you want to get better than that that things can get complicated.
Getting the angular accuracy for that level of position accuracy is a different matter. 50cm lateral movement at a range of 5m would require an angular accuracy of sin-1(0.5/5) = ~5.7 degrees.

The DW3000 data sheet gives range accuracy as +/- 6cm and angle accuracy as +/- 10 degrees. You can maybe play games and use some averaging to improve things but on paper getting the angular accuracy you would need for that position accuracy may be tricky.

For this application you would get range by using TWR, a reply is sent a fixed amount of time after the initial message is received. The initiator then measures the total time, subtracts the fixed delay and so can calculate the average time of flight of the messages.

Two antennas gives you a 180 degree ambiguity which 3 antennas would fix. Probably have a primary antenna and then two secondary ones at 45 degrees to each other, you then use an external RF switch to toggle between the two secondary antennas and so solve the ambiguity.