Making the schematic

After a bit of a marathon of routing the schematic, I have the first version:

There is no doubt that there are tons of mistakes in it, a few that I am yet to find. That being said, here are my top priorities to fix:

• Filtering the power supply. First of all, do we use a 400KHz or 2100KHz switching frequency? Something in between? Are higher or lower frequencies better for RF work? My primary concern is that some harmonics/ripple will get into the DRA818's power supply and ruin the sensitivity. Even worse, they could get past the AMS1117, since it doesn't filter out very high frequencies, and get into the LNA power supply, possibly amplifying them into the RF input.
• Add more test points. While I initially didn't like blotting test points all around my board like it was a bad case of teenage acne, I've come to realize their usefulness. We need more test points!
• How to we add more user interface? For example, how does the user set the transmit/receive frequencies? It's very possible to just hardcode them in firmware (which I am tempted to do, since I'm the person writing the firmware anyways, and every gram is precious on a high altitude balloon), but could there be a better way?

That's all for now. Let me know if you have any comments/solutions/ideas!
Schematic in PDF

Working principles

Thinking about exactly how this can work, and what parts to use!

Goals:

• Repeat VHF to UHF FM signals
• Be configurable to different frequencies, power levels, etc.
• Identify itself on the air with morse
• Have a maximum transmission limit of ~15 seconds to prevent overheating

I'm planning to use a DRA818V module for receiving and a DRA818U module for transmitting. Each will be configured via UART to set frequencies, CTCSS, etc. I might also add a GPS receiver on board to transmit it's location once in a while.

I also need to think about the type of microcontroller to use. I'm leaning towards an STM8, but I've never worked with STM8s before and only a bit with STM32s. STM8s are less pricy but also appear less well supported. Either way, my only requirements are having 3x UARTs and about 10 GPIO pins. Maybe an STM32 will be better.

In terms of PA, maybe it's worth reverse engineering the DRA818 module? Need to think about this