Hi all. I recently came across the pyCubed project. It’s a opensource board which includes the OBC, EPS and COMMs (LORA), admittedly doing some simplifications and not including a PC104 connector. It was used in the V-R3x mission (triple 1U Cubesats), launched 24.Jan.2021.
I think it’s worth a look, and I’d be curios about your input.
Open Source software: Check!
Open Source hardware: Check!
They have made a really good effort with this, and I commend them for releasing it as they have. However…
Open Source Radio Modulation: Nope. LoRa. Again.
As has been discussed in another thread on this forum, a lot of us feel that LoRa doesn’t fit into the open source space ecosystem. It’s a proprietary modulation, and the SDR-based implementations which do exist (e.g. gr-lora) had to be reverse engineered and perform poorly as a result. Then there’s the question about patent infringement, and how Semtech would respond to an open source, reverse engineered implementation of their modulation.
LoRa appears to be the ‘in thing’ at the moment, and allows satellite designers to ride the IoT buzzword train. But there are alternatives available which don’t require the use of proprietary modulation, and would allow for the use of SDR-based ground-station networks (like… SatNOGS). The AX5243 (successor to the AX5043) is one such example of a radio chip that will do this.
Shameless plug: our open implementation for a COMMS board around AX5043 https://gitlab.com/librespacefoundation/pq9ish/pq9ish-comms-vu-hw/
Thanks for sharing @cclaudiu.
The project looks great (for education purposes) despite of the LoRa discussion. It’s Open Source HW&SW, it has quite a lot of documentation, code is Micropython… Of course there are a few things I would like to happen such as different comms, better code maintenance or docs on a different platform (now in Notion)… But overall I think it’s worth a look.
Thank you all for your feedback!
I agree with the view on the (closed) LoRa. My take on the RDM98W was that it’s a “turn-key” COTS module, capable of “FSK, GFSK, MSK, GMSK, LoRa and OOK modulation” (so not necessarily LoRa), and by using the free bands we could bypass the need for IARU registration (maybe?!). Not to mention simple/cheap groundstations with similar hardware. Thank you very much vk5qi and pierros for providing an alternative.
A key point about the pyCubed is that it integrates COMM, EPS, OBC, GPS (no ADCS) into a single board, greatly saving space for a 1U CubeSat. Most of the current solutions have separate boards for each of them. I know that some redundancy or safety mechanisms have been omitted, but maybe it’s a good tradeoff for some missions.
‘Free Bands’ are not ‘free’ worldwide, nor are they consistent.
E.g. the US’s ISM band is 902-929 MHz. In Australia, it’s 915-928 MHz, in Europe it doesn’t exist (they have a band at 868 MHz instead). So, those VR3X sats that launch recently transmitting on 903 MHz are actually transmitting on a commonly used mobile phone band when not over the US.
Even the 434 MHz ISM band is not universally recognised.
Finally, there’s the issue of licensing - none of these ISM bands are actually approved for space use - you still need a license.
Would be great if open community comm libs like RadioLib https://github.com/jgromes/RadioLib/issues/194 added support for AX5243 since some PyCube VR3X sat decodes (LoRa) are done through open community cubesat software like TinyGS software https://twitter.com/tiny__GS that uses RadioLib on the back end. TinyGS supports both LoRa and FSK modulation among others. For example, Norby sat alternates between transmitting LoRa and GFSK beacons every other minute, SatNogs gets the GFSK decodes, and TinyGS gets the LoRa decodes, but TinyGS could get both decodes with more support for open chips like AX5243 etc. The reason that would be nice is that currently, many attempt to use LoRa chips in their non-LoRa modulation for cubesat decodes, and AX5243 chips offer an open option.