Forums › SM Series Discussions › Compressed IQ stream
- This topic has 6 replies, 4 voices, and was last updated 4 years, 7 months ago by Anonymous.
vanderbasaranParticipantvanderbasaran March 9, 2018 at 12:23 am
In the API document it is mentioned that Compressed IQ streaming is not implemented yet. What is the plan here?
AndrewModeratorAndrew March 9, 2018 at 9:03 am
I would expect this functionality within the next 6 months.
Is it possible to use Spike and the compressed IQ stream to analyse the 2.4 GHz WiFi Band (80 MHz Bandwidths) in Time Domain (Zero Span) to measure the length and the channel of Bluetooth packets ?
Justin CrooksModeratorJustin Crooks August 20, 2018 at 10:32 am
Preliminary tests on the compressed I/Q streaming indicated that we would end up with fewer channels than we had hoped for, and the interface would be prohibitively complex for most users, so we moved away from it, towards just adding additional memory for full 160 MHz I/Q captures.
Next year, we will have units available with this expanded memory. With video or frequency mask triggering, it should be straightforward to capture and analyze Bluetooth packets.
Instead using the the zero span to measure the duration of a packet what is the limitation of using the waterfall spectrum. Is it possible to get 10 us time resolution for a 80 MHz wide signal and perhaps 10 to 80 channels ?
AndrewModeratorAndrew August 21, 2018 at 9:09 am
Good question, and it’s a bit tricky to answer. Our waterfall plot currently is only available in the sweep mode. Each sweep is plotted on the waterfall and generally each sweep is going to take 1ms+ depending on the analyzer and settings. So that would be your minimum resolution.
To get 10us resolution in a waterfall plot, we would need to introduce the waterfall plot into the zero-span mode (or some other I/Q mode) and perform overlapping FFT’s on the IQ data. We don’t have anything like this in Spike right now.
As Justin mentioned, our max bandwidth IQ device is 40MHz right now with the SM200A, or 27 MHz with the BB60C. Both are programmable and this type of resolution is possible if you were willing to program and work with the I/Q data yourself using our API.
We are working to increase the I/Q bandwidth of the SM200A to 160MHz for up to 1 second captures. This is still in development.
The expanded memory option would be very useful but I still believe that you should also think about to capture 10 us packets with 250 MSPS, perform a overlapping FFT and transfer the spectrum in real time over the USB3. So finally there is the possibility to analyze long records with a very fine time resolution. The fine time resolution is needed because there are short packets (minimal 44 us duration for Bluetooth 5 packet) in the 2.4 GHz Band. For the persistence display you do similar things in the FPGA, but instant to accumulate – transfer the spectrum every 10 us.
You must be logged in to reply to this topic.