Cardiff Microwave Roundtable - 16th March 2019

[MW0LNA]

I'm very happy to share that there will be a one day UK Microwave Group Roundtable in March (yes! Just two months away!) this year. The Cardiff University Amateur Radio Society is organizing the event which will be hosted at the university with a dinner nearby in the city center. We have a number of talks already confirmed ranging from Noel G8GTZ talking about ATV, myself on Digital Pre-Distortion with SDRs, and two talks about microwave power amplifier designs.

Please mark your calendars! We hope to see you there!

https://www.cardiffars.org.uk/roundtable

There are a number of talks during the day for which we are looking for speakers. If you are interested in presenting a project or topic please get in touch with myself or officers@cardiffars.org.uk. The spaces are 30 minutes long by default but we're very willing to move that around as best fits your presentation.



There will be space for displaying projects and a test equipment station for measuring components and systems. Please let us know if you'd like to display something or have equipment you would be willing to bring.



Best regards,
Derek Kozel & the Cardiff University ARS
MW0LNA & GC0CDF

[G4GUO]

Will it be streamed or recorded?

- Charles

[MW0LNA]

Hi Charles,

We will recording any speakers who agree. Setting up streaming is a bit more work than we're able to put in this year, but hope to do so for next year.

[G4GUO]

Hi Derek,

I am mainly interested in the DPD material as that is what I am currently working on, overheads and an audio recording would be of interest.

Thanks

- Charles

[MW0LNA]

The DPD material will certainly be available, but I suspect that it will be of only little interest to you. I'll be talking about volterra series and memoryless polynomial based DPD. I am working with someone else who plans on implementing more advanced DPD algorithms in a GNU Radio block so maybe some of that will be ready by March and I can also talk about them.

[G4GUO]

Hi Derek,

That should be interesting as the simplified Volterra approach is something I have looked at.
I tried to give a simplified talk about DPD at CAT 18 and I have read Dr John Wood's book
"Behavioral Modeling and Linearization of RF Power Amplifiers". Wood has put one of
his talks on YouTube if you ever have difficulty sleeping.

I am looking at writing a DPD IP block that I can include in the PL fabric of a Zynq device (Zedboard).
I have just been talking to some people at Analog Devices so I am pretty confident the FMCOMMS4
hardware will be suitable, if I use the TX Monitor input rather than the normal RX input the monitor
channel and transmit channel use a common synth.

At the moment I am not sure whether I will use a Polynomial filter or the method that Altera
describe in their appnote (LUT and Polar maths).

I did write a C implementation that used a poly filter length of 7 and a memory 3 deep but I have not managed
to get that to work yet. The coefficients generated by my linear algebra were unstable on real signals
but worked fine on test data.

- Charles