Repeater Project: Zero IF NCO Up-Convert

Now that I’ve got a simple 806MHz repeater working it seems like I need to up convert it to 826.5MHz. Can this be done in Pothos or GNU radio ??? - I can’t see any blocks that specifically do this. Otherwise I’m gonna have to trawl through settings in LimeSuite - no luck so far.

I have been trying to follow this Lime Microsystems tutorial:

I’m convinced that the tutorial itself is pretty good EXCEPT that somebody in Lime microstsyems has deleted the essential repeater.ini file from GitHub and the guy who wrote the tutorial seems to have disappeared. Has anybody found this guy or the .ini file yet? :worried:

Here is an excerpt from the tutorial:

"LimeSDR contains LMS7002 transceiver, it is MIMO - it means it has two channels, each channel has RX and TX (2x RX and 2x TX in total). And it contains one PLL for both RX in Channel A and B and one PLL for both TX in Ch A and B.

As different Mobile Operators are using different frequency bands and TX to RX separation will be also different, we will use universal configuration:

Channel A we will use to receive a signal from the Base Station and retransmit to Mobile Phone and Channel B to RX from Mobile Phone and TX to Base Station.

Now when we know frequencies (from the previous step), we can calculate Centre Frequencies and offsets:

Centre Freq for RX from BS to Phone: (791+821)/2 = 806 MHz (30 MHz bandwidth)

Centre Freq for TX from Phone to BS: (832+862)/2 = 847 MHz (30 MHz BW)

To keep configuration universal for all LTE bands, we will use only one PLL (TX PLL), that will set LMS7002 Centre frequency (LO), it will receive and transmit at this frequency:

LMS7002 LO: (806+847)/2 = 826.5 MHz

It means received IF (demodulated low freq component, without RF) signal frequency (Fin - Flo) will be from 791 - 826.5 = -35.5 MHz to 821 - 826.5 = -5.5 MHz (-20.5 MHz centre)

In RF terminology, when you receive not at the center frequency, it is called “Low IF”, so now received bandwidth became (826.5 - 791)*2 = 71 MHz (wanted signal is in 30 MHz BW).

“Zero IF” is more efficient for digital signal processing, because you can work with narrower bandwidth (compared to “Low IF”) and use lower sample CLK or get more taps for the digital filter. In this Demo we will not digitally process signal - just connect RX digital output to TX digital input (digital loopback).

  1. To work in “Zero IF” (when you receive in the middle of the band), we will shift it by +20.5 MHz to 0 using LMS7002 internal NCO (numeric controlled oscillator), it is called upconvert. So now our signal will be from -15 to +15 MHz (0 MHz center).

  2. Then enable Digital loopback.

  3. Now, when the signal is in TX chain, use NCO to shift it by -20.5MHz (to original), it is called downconvert. And the signal will be transmitted to mobile phone

  4. Signal from mobile phone will be received by Channel B RX (847 MHz RFin), it means 20.5 MHz NCO downconvert for RX and upconvert for TX.

  5. Signal from mobile phone is strong, so Ch B RX gains should be low.

Summary: calculate LO, calculate how much you need to shift by NCO. Channel A and B have same settings, except NCO upconvert/downconvert and gains."

I thought project makers had to put all the project info on hackaday. Maybe hackaday has a backup copy???

Did you check github for .ini files? Having lots of example .ini files would really help users getting a start.

I think that you can put as much info on Hackaday as you like. It works on a progressive basis, which is quite nice, so if you’ve got a tricky project you can show progress in the ‘logs’ section and maybe people will make useful suggestions in the comments section.

Yes, I checked and re-checked the GitHub for the files and every time I checked, they still were not there!

Maybe the author took them down because they didn’t want to support them with newer releases of software/hardware. I guess we’ll never know but I think D. Webster might. :wink:

1 Like

Very possibly … But it does not matter now … Now that I have some results I’ve lost interest in the whole alien conspiracy thing.
PS. I think my phone is probably on band 3 but need to plant potatoes before it rains.

[quote=“TegwynTwmfatt, post:5, topic:1125, full:true”]

some results I’ve lost interest in the whole alien conspiracy thing.
[/quote] It’s vast.

So your band 20 filters won’t work for you? You need to make band 3 filters?

My potatoes seem to grow back some on their own. I think because I didn’t actually take them all out of the ground last time.

Meant to mention this, it is meant for fast prototyping. I browsed the docs and tried one example. Seems simple enough and it’s updated. To me it looks easier than to use LimeSuite. I think it is python2 only FWIW.