I have recently been investigating how to extract TDOA from CIR for location estimation. I refer to the Snaploc: An ultra-fast UWB-based Indoor Localization System for An Unlimited Number of Tags published by Bernhard Großwindhager of the Department of Electrical and Information Engineering, Graz University of Technology, Austria at the 2019 IPSN meeting, and attempts to reproduce his experiment. In this paper, it is mentioned that a reference anchor is used to initiate the communication process, and the other anchor points send a response message at a fixed time interval. During this process, tags only monitors and extracts the CIR information.
I tried to repeat the experimental process by 3.3 Delayed Transmission in DW1000 User Manual. I assigned different addresses to the three anchors. After the anchor received the information of the reference anchor, the Delayed Transmission would start after 500us (considering reasons such as SPI communication delay) + anchor point address * 128ns delay.
It was supposed that a complete CIR image of 16MHz mean PRF would contain 992 sampling points, each of which was 1ns, so I should have been able to capture responses from at least two anchor in a single image, but it didn’t work out.
The CIR I captured is shown below.
During the experiment, I could judge that the three anchors had received and sent response information through the flashing of the LED light.
So my question is:
Did I repeat the experiment in that paper correctly?
Will 128ns interval cause signal collision or there are other reasons that DW1000 cannot capture UWB signals sent by multiple devices within a CIR cycle?