Re: [44net] NPR (New Packet Radio) : new firmware with 56kBaud and 120kBaud

Hello Guillaume, This is fantastic news and thank you for working on this to allow for other regions to give it a try! Few questions if I may now that this is a reality to try: - I see in the advanced guide that the system "IDs" itself every 2-6 seconds but what mode is that done in. In line with the symbol and spectrum limitations that the US amateurs have to deal with, IDing is also something to deal with. Is your system able to send non-NPR transmissions? http://www.arrl.org/part-97-text --> 97.119 Station identification . There is some discussion about this on the Hackaday comments section but there doesn't seem to be a conclusion about it. - In the current firmware, I don't see any mention of a watchdog if the system crashes. Is it possible to add one? - I see in the docs and videos the details about client to master but with this design, can say client #1 transfer data to client #2 via the master? Can client #1 automatically wake up client #2 and send data (assuming there is an IP server to connect to? - Your advanced document mentions that due to weakness in the FEC algorithm, maybe using modes "11, 20, or 21". I see there isn't any mode "10" for the 56kS/s symbol signaling rate. Is this a limitation of the modem chip? - Your Introduction PPT mentions a upcoming kit. Any ETA and cost on that? On the Hackaday page, you mention you would ship to the EU but does that mean you WON'T ship to the US? Making at least the PCBs, sheet metal and a parts BOM would be a huge help for us in the US. - Interference on 433Mhz is always going to be a challenge but do you think it would be possible to add a "scanning" feature to the NPR system so it could go out and listen to X frequencies for Y seconds and come back with a report if those frequencies are quiet? This is something that Wifi has had on it's "channels" and it can be hugely helpful. In an amateur radio deployment, I don't that we would ever want a auto-channel selection feature but being able to initially listen and pick a frequency would be a huge help. The only other option here would be for the operator to deploy an SDR and listen in. - With the lower bandwidth design, do you think enhancements could be made to support full duplex using two radios? Aka.. each transmitter could use multiple time slots for higher performance? I do understand how that makes the overall setup a lot more complex. - You mentioned that NPR doesn't support IPv6 which is ok for small deployments but there isn't any mention of MTU. I assume this system supports a RAW MTU of 1518 bytes but maybe it can support smaller or larger MTUs to better deal with the BER rate of the connection? - I see on the Hackaday page that it looks like the expansion from 7 clients to 15 clients might need different hardware. Would it be a good idea to socket the STM board so the upgrades could be "plug and play"? - I see you have email lists other community links posted on the Hackaday page. Would be good to add those links to the end of the Introduction PDF - Maybe this Q&A and some of the other points from the Hackaday comments section would all be well suited to start a FAQ? Anyway, thanks again for your efforts here and making it available to the world! --David KI6ZHD On 06/16/2019 06:30 AM, f4hdk wrote: