Hello,
I have a project where I want to locate a person in an outdoor area. I am using MDEK1001 modules, specifically 4 anchors and 1 tag to locate. All the tags are placed in a 5x7m rectangle.
During the tests, a good location has been achieved, but unwanted phenomenon have been detected. When there is a person next to the module, the location can be lost. Also when there is a car nearby. I have made several tests increasing the number of tags to see if it improves but it has not.
What possible improvements or tests can I do to minimise this effect?
Thank you.
Hi @ecastro ,
In-body attenuation can cause huge losses at high frequencies and therefore a broken link. You may try to use low-frequency channels to improve that. However, since two-way ranging is a time of flight based measurement, the speed of the propagating wave would be different in the body compared to air and result in inaccuracy in your range measurement.
Furthermore, reflections from the car can also cause destructive interference from the reflected path. If this is the case, the only solution would be to remove the reflector from the measurement field.
I hope this information is useful for you and let us know about your progress!
Kind regards,
Emre
Hello Emre,
Thank you very much for your quick response.
I have been searching but I can’t see how to use the low frequency channels, could you guide me a little?
On the other hand, is it possible to replace the antenna to mitigate this effect or would I still have the same problems?
Thanks again.
Best regards.
Changing the channel may help a bit. Changing the antenna may help a bit. But ultimately if there is something RF absorbent (a person) or reflective (a car) in the path or close to the antenna it’s going to have an impact.
If you can move the antenna to above peoples heads then it will be a lot better but that may not be practical.
Beyond that you get into more complex position algorithms, either using more anchors in the position calculation so that one being blocked has less impact or trying different combinations of anchors and looking for inconsistencies to indicate which lines of sight are blocked.
Hi Ecastro,
When Emre mentioned that you can try low frequency channel there should be an option on the MDEK to change the channel to channel 5 (6.4896GHz), one of our software guys @carlos.silva may be able to help on how exactly you can change the channel in software. Hardware you will just need a micro usb to connect to the side of the MDEK and usb to connect to a desktop or laptop with and terminal application e.g. Tera Term or Putty. The antenna provided on the MDEK should be good for None Line of Sight (NLOS) applications infact so there won’t be any need to change it.
If Carlos can help you communicate and change the channel there may be other radio configurations to improve NLOS.
Thanks,
James
Hi @AndyA , @jamesdoyle ,
Thanks again for your help.
If @carlos.silva could tell me how to do it, I could do a comparative test between the two frequencies to see the result.
Best regards.
Hi @ecastro !
I’m checking the software available for your modules, and I’ll return later today.
Kind regards!
Hi Carlos,
Thank you very much. I look forward to your reply
Regards.
Hi @ecastro !
I’m still checking this.
Even though the DW1000 chip does support channels 1-5 and 7 I couldn’t find software supporting it. Checking the DWM1001 System Overview, which includes the MDEK1001, points that the UWB Channel parameter as “Channel 5 (6.5GHz)” and “Fixed”.
I’ll try to find other software options but, at first, setting the channel is not supported. Maybe the antenna in the module is not rated to the other channels.
Kind regards!
1 Like
Hello @carlos.silva ,
What bad news… I’m waiting to see if you find something that can improve the situation.
Thank you very much.
Best regards.
Hi ecastro,
The antenna on DWM1001 is ch5 only.
See this thread:
1 Like