WEBVTT 00:00.000 --> 00:13.000 All right, since euro on time, so far we, so it's 9 a.m. 00:13.000 --> 00:18.000 Thank you for joining early in the morning, the SDR Devroom. 00:18.000 --> 00:23.000 So it's going to be a full day of talk today, fully packed. 00:23.000 --> 00:28.000 So hopefully we can keep this schedule and if you want to switch from one room to another, 00:28.000 --> 00:33.000 we try to keep the schedule as planned initially. 00:33.000 --> 00:38.000 Quick word of introduction, software radio and DSP. 00:38.000 --> 00:42.000 Why do we have this title this year? 00:42.000 --> 00:47.000 Every year, there's, for the last few years, there's been a couple of submissions to 00:48.000 --> 00:50.000 hand radio and software-defined radio. 00:50.000 --> 00:56.000 And unfortunately, the, for the organizers, had a hard time figuring out that that's two different fields. 00:56.000 --> 01:01.000 It's got radio in it, but this year to try to differentiate 01:01.000 --> 01:08.000 discrete time, complex sample, IQ processing from software control radio. 01:08.000 --> 01:13.000 We added the digital team processing and, well, this year, that's the one that was selected. 01:13.000 --> 01:16.000 So thank you for joining. 01:16.000 --> 01:24.000 As I was not so kindly reminded, as Non-Hem Radio, we are not supposed to broadcast. 01:24.000 --> 01:28.000 At least those of you who do not have a ham radio license. 01:28.000 --> 01:34.000 So I just wanted to remind you that in case you wish to broadcast a signal, you must comply with 01:34.000 --> 01:40.000 industrial scientific and medical band regulations, which I put back for you here. 01:40.000 --> 01:42.000 So we've got some HF. 01:43.000 --> 01:48.000 We've got some UHF, we've got some S-Band, we've got some C-Band. 01:48.000 --> 01:58.000 So we've got plenty of ISM non-licensed bands to broadcast on and anyway, there's so many useful signals out there. 01:58.000 --> 02:06.000 So that we have so much to receive already to decode that there's plenty of signal to analyze anyway. 02:06.000 --> 02:14.000 The power levels are all defined limitation or updated power and duty cycles are provided in the IT regulation. 02:14.000 --> 02:18.000 So I'm not going to read the whole hundred and sixty or something pages of ITU. 02:18.000 --> 02:24.000 So start already by broadcasting only the ISM bands, then we comply with regulations. 02:24.000 --> 02:30.000 And I'm going to show you actually how you can broadcast below the terminal noise, which means that even in all the other bands, 02:30.000 --> 02:37.000 you all have the terminal, the noise limit, where you won't be detected anyway. 02:37.000 --> 02:43.000 Another topic that I really enjoy about first them is not being a so-called scientific conference, 02:43.000 --> 02:50.000 where actually people don't just want to show off results, but they really want to share how they actually result and how they did it. 02:50.000 --> 02:56.000 So I checked on the first them, SDR dev room, all the contributors provided a GitHub, 02:56.000 --> 02:59.000 so they took the reproduced very experiments. 02:59.000 --> 03:06.000 And during the Christmas occasions, as I was reading, Dan Wong's book, comparing the Western politicians, 03:06.000 --> 03:11.000 with their lower background and the Chinese politicians, with engineering background and high-tech, 03:11.000 --> 03:14.000 probably much the same issues. 03:14.000 --> 03:22.000 I was attracted to this paragraph where they highlight why Shenzhen is so in the Hong Kong, of course, is so successful. 03:22.000 --> 03:27.000 And they argue that that it's a community of engineering practice, 03:27.000 --> 03:35.000 whereas in the Western Silicon Valley, we're obsessed with creating new stuff and new invention. 03:35.000 --> 03:43.000 Shenzhen is more about viewing blueprints and technological progress as training of various scientists, 03:43.000 --> 03:46.000 manufacturers, 20 technologies, people, and process knowledge. 03:46.000 --> 03:55.000 And I believe that at this gathering, we have a few people in Europe that are still in this mind of engineering as a way of life, 03:55.000 --> 03:59.000 as a rather than just a day job for getting money. 03:59.000 --> 04:05.000 So I really enjoy organizing this presentation with these people. 04:05.000 --> 04:11.000 So a few highlights, this is a very personal summary of this year, of the last year, 04:12.000 --> 04:17.000 which has no claim of exclusivity. It's just a few ideas I had from this year. 04:17.000 --> 04:24.000 We've received, you might have heard, a Blue Ghost Mission, which landed on the Moon, 04:24.000 --> 04:31.000 and so it was sponsored by the Italian Space Agency and the National Space Agency of America, so the NASA. 04:31.000 --> 04:40.000 And what they did is they recorded global navigation satellite signals from the Moon, on their path to the Moon and on the Moon. 04:40.000 --> 04:52.000 And they released the IQ data on the Sun website where you will find on the general record here, 04:52.000 --> 04:54.000 all the data they collected. 04:54.000 --> 04:59.000 And when they released the data, they had a meeting in Rome for two days. 04:59.000 --> 05:04.000 If you're interested, I didn't find the release of a recording of the video conference, 05:04.000 --> 05:07.000 but I do have a copy on my computer, if you're interested. 05:08.000 --> 05:13.000 The first day was mostly political stuff from ESA, but the second day was really technical about the Blue Mission. 05:13.000 --> 05:22.000 And they actually released the data and they opened a challenge asking, in one year, what are people going to achieve with these IQ streams? 05:22.000 --> 05:28.000 So you can try by yourself, if you're familiar with or even if you're not familiar with Genesis processing, 05:28.000 --> 05:33.000 that's a good way of getting attracted to this topic. 05:33.000 --> 05:40.000 Another highlight for me, I've been waiting for nearly 10 years for a Spaceborne satellite, Nissan, 05:40.000 --> 05:44.000 between the Indian Space Agency, Israel and NASA. 05:44.000 --> 05:47.000 And it's been launched in July, 2025. 05:47.000 --> 05:54.000 And this is an L-band, so 1.2GHz and S-band 3GHz satellite radar. 05:54.000 --> 05:59.000 And I'm going to show you, actually, if we can receive the signals, and then you can actually have a lot of fun. 05:59.000 --> 06:04.000 And at the moment, there's a whole collection of Spaceborne radars. 06:04.000 --> 06:08.000 There's a couple of companies, Capella, Umbra, just fantastic. 06:08.000 --> 06:13.000 If you read the BBC News, you're going to see a few of these pictures illustrating the BBC News. 06:13.000 --> 06:20.000 They show 25 centimetre resolution radars, so all the weather, all elimination from 80 kilometers away. 06:20.000 --> 06:22.000 Just fascinating. 06:22.000 --> 06:27.000 This exercise is an origin project of Spaceborne radars. 06:27.000 --> 06:31.000 There's a whole collection of Spaceborne radars broadcasting this information. 06:31.000 --> 06:36.000 And again, Nissan, similarly to Sentinel-1, and to Lugar, 06:36.000 --> 06:42.000 are claimed to publish the raw IQ streams, the little zero data sets. 06:42.000 --> 06:45.000 At the moment, I have not seen data. 06:45.000 --> 06:51.000 I think this week, I saw a few first files being uploaded on the Alaska satellite facility ASF. 06:51.000 --> 06:56.000 You need an account, but it's free of charge, and I think you can give a dummy email address. 06:56.000 --> 06:58.000 They're not going to verify. 06:58.000 --> 07:05.000 Unfortunately for us, also, beyond these launches, I think many of you might have started 07:05.000 --> 07:09.000 HDR with the low Earth orbital satellite from NOAA. 07:09.000 --> 07:11.000 And that's finished now. 07:11.000 --> 07:17.000 In August, the last broadcast of the NOAA weather analog satellite was switched off. 07:17.000 --> 07:25.000 And at the end of since the 1970s of continuous observations of Earth using the analog signals from NOAA. 07:25.000 --> 07:33.000 So now we left with, well, to the best of my knowledge, the M2 versions, CCSDS compliant satellite, 07:33.000 --> 07:40.000 which, again, makes the entry level of HDR much steeper because we used to have an analog signal. 07:40.000 --> 07:45.000 You just put a white band FM receiver, and you send this on your soundcard, you add something to listen it. 07:45.000 --> 07:55.000 Now you need a 2PSK decoder, and then you get JPEG images, and you need to assemble the JPEG images to meet to one year to get to figure out how much M2 was working. 07:55.000 --> 08:02.000 So that's an over challenge, and actually my colleague, to my left hand, spent a bit of time decoding the latest M2M2M, 08:02.000 --> 08:07.000 because they defer each of the M2D for a little bit from the previous one. 08:07.000 --> 08:13.000 So you need to adapt the decoding procedure. 08:14.000 --> 08:21.000 One of the highlights for me, we have one new radio developer, where the CEO is on the back. 08:21.000 --> 08:26.000 One of the highlights for me is new radio is table. 08:26.000 --> 08:31.000 It hasn't changed forever, it doesn't change API, it hasn't changed configuration file. 08:31.000 --> 08:39.000 For a second year in the rule, now I think we have the same new radio version on all distributions, on all operating systems. 08:39.000 --> 08:51.000 In January, I taught in the Paris University 15 students, 14 of which were running Windows, unfortunately, this means that still 90% of the French students are running property reporting system, 08:51.000 --> 08:57.000 and all of them, to the best of my knowledge, managed to install the new radio on their Windows computer and rental applications. 08:57.000 --> 09:08.000 So that's very nice, because it means that instead of keeping on repairing things that used to work, which seems to be a habit in open source software, we now can focus on getting worked on. 09:08.000 --> 09:12.000 There are a few of a few of the work that was achieved. 09:12.000 --> 09:21.000 This is one example, again, with Toma Lavaren, using a new Windows scope for gathering SDR signals, I mentioned this yesterday at the both meeting. 09:21.000 --> 09:30.000 So collecting data from a new Windows scope, of course, cannot be continuous, because your data rate from the new Windows scope to the computer is finite. 09:30.000 --> 09:37.000 But yet, many applications, you can just gather burst of information, process of information, and then gather the next burst. 09:37.000 --> 09:43.000 Whether it is for radar, where we have a reference signal in our surveillance signal, or in this case for the oscilloscope. 09:43.000 --> 09:51.000 So we were collecting 100 megabit per second oscilloscope streams and analyzing the data. 09:52.000 --> 09:55.000 Just for a little story, we were looking at this with Toma. 09:55.000 --> 10:02.000 And at some point, I'm sure you all know that the Ethernet header is 800, so we were looking at all these frames with 800. 10:02.000 --> 10:09.000 And at some point, as we were looking at the continuous stream, we had a 30 Ethernet type 887. 10:09.000 --> 10:14.000 Anyone familiar with what this is? 10:14.000 --> 10:19.000 No, we're not that advanced. 10:19.000 --> 10:22.000 Actually, it's funny, it's a PTP header. 10:22.000 --> 10:31.000 And the reason I mentioned this is that you'll see that a couple of presentations today are focused on the white rabbit, synchronization system, which is an enhanced PTP. 10:31.000 --> 10:34.000 We're going to talk about it and how it relates to software different radio. 10:34.000 --> 10:40.000 And it happens that in the lab, I have a white rabbit network, which was supposed to be isolated from the rest of the network. 10:40.000 --> 10:48.000 And actually, you can use an oscilloscope to debug your IP tables errors, because, obviously, we have PTP frames leaking in the network. 10:48.000 --> 10:54.000 And this we discovered when we're looking at our oscilloscope outputs. 10:54.000 --> 11:01.000 As you are for topics, you all know that the radio runs flawlessly on embedded systems. 11:01.000 --> 11:06.000 My colleague, Gwynel Goavec, has maintained a new radio functional build route. 11:06.000 --> 11:14.000 A field balancer has kept open embedded the recipes for new radio working on open embedded. 11:14.000 --> 11:21.000 So, we also have a new radio functional on your headless embedded systems. 11:21.000 --> 11:26.000 We did some work with voice check on GRM17. 11:26.000 --> 11:30.000 Basically, I just wrapped his C library into the new radio framework. 11:30.000 --> 11:32.000 He tells me it's all functional. 11:32.000 --> 11:39.000 I must admit, I did not check, but it's also another final application of the new radio. 11:39.000 --> 11:47.000 And we're going to have a first presentation this morning about various digital communication schemes. 11:47.000 --> 11:49.000 Next to M17. 11:49.000 --> 11:53.000 So, it's going to be a discussion about digital communication. 11:53.000 --> 11:57.000 Finally, because now we have a stable framework to work with. 11:57.000 --> 12:06.000 Not only do we have the opportunity to work more on OT and out of tree blocks rather than keeping on repairing broken frameworks. 12:06.000 --> 12:10.000 But also, there's a huge improvement on the documentation. 12:10.000 --> 12:20.000 Barry Degen and a couple of his friends on the chat that you can join on the documentation. 12:20.000 --> 12:29.000 The channel has been pulling together their activity about trying to give documentation about each and every block. 12:29.000 --> 12:32.000 And giving examples on how to use these blocks. 12:32.000 --> 12:34.000 So, really try to have a look. 12:34.000 --> 12:37.000 I had a quick look at the statistics. 12:37.000 --> 12:42.000 It's more than 130 entries now with about 10 files per entry. 12:42.000 --> 12:48.000 So, it's a great entry point for getting started with a new radio and DSP in general, 12:48.000 --> 12:56.000 because it gives you the opportunity of getting familiar with the practicals of each tool processing framework. 12:56.000 --> 13:00.000 What's going to happen in the year? 13:00.000 --> 13:04.000 If I look at last year, usually we have a few governing. 13:04.000 --> 13:05.000 So, here you are at one of them. 13:05.000 --> 13:09.000 The software you find radio first them at the beginning of the year. 13:09.000 --> 13:14.000 During the summer, we should have the software you find radio academy. 13:14.000 --> 13:17.000 Organized by Marcus. 13:17.000 --> 13:22.000 In a pretty shuffin, always end up doing a fly. 13:22.000 --> 13:26.000 Last year, there was no European new radio days. 13:26.000 --> 13:28.000 It was a bit difficult to organize. 13:28.000 --> 13:33.000 The previous year was lacking participants and especially speakers. 13:33.000 --> 13:35.000 So, we decided to postpone to this year. 13:35.000 --> 13:42.000 So, Silmore, organizing Lyon, an informal gathering of enthusiasts. 13:42.000 --> 13:48.000 And hopefully next year, sorry, this year, we should have by the end of the year November or December. 13:48.000 --> 13:50.000 It's not yet planned. 13:50.000 --> 13:53.000 A European new radio day in Western France in Britain. 13:53.000 --> 13:55.000 It's been organizing rain. 13:55.000 --> 14:03.000 So, hopefully you can keep your schedule slot to join in Western France European new radio days. 14:03.000 --> 14:07.000 This year, we had the wire conference in the Netherlands. 14:07.000 --> 14:15.000 A very enthusiastic group of hackers meeting on all topics, including software in front radio. 14:15.000 --> 14:20.000 There was an M17 meeting in Poland organized by boycheck. 14:20.000 --> 14:25.000 And, of course, in September, our own September will have the new radio conference in the US. 14:25.000 --> 14:27.000 So, that's going to happen. 14:27.000 --> 14:30.000 This schedule for already scheduled and already announced. 14:30.000 --> 14:36.000 There is a call for contribution open for the new radio conference in the USA. 14:36.000 --> 14:44.000 No, I went in wrong direction. Sorry. 14:44.000 --> 14:49.000 Right. So, that concludes my highlights from 2025. 14:49.000 --> 14:54.000 So, just as a quick recap of today's program, which is fully booked, 14:54.000 --> 14:59.000 we start with a contribution from the Open Research Institute, 14:59.000 --> 15:05.000 which will actually be presented by Evernist about space and terrestrial digital communication. 15:05.000 --> 15:12.000 Then, we have a few discussions about web-based software and radio communication. 15:12.000 --> 15:20.000 I'll be sharing how I've been using the TWSDR signals to decode some of the very low frequency time transfer signals, 15:20.000 --> 15:23.000 including Loran and Eloran. 15:23.000 --> 15:30.000 How to collect data from China, from Russia, from Saudi Arabia, through the TWSDR. 15:30.000 --> 15:33.000 And, we have the WSDDR presentation. 15:33.000 --> 15:39.000 Hopefully, it's going to be a remote presentation because the offer of a speaker was not able to get a visa. 15:39.000 --> 15:43.000 So, we'll try to make sure that the online presentation goes well. 15:43.000 --> 15:47.000 In between to get ready, we have a surprise for you. 15:47.000 --> 15:52.000 You're going to see an amazing signal, hopefully, broadcast live from deep space. 15:52.000 --> 15:56.000 So, that's going to be a teaser for the presentation. 15:56.000 --> 16:00.000 The last two before presentation about SDR for radio telescopes. 16:00.000 --> 16:05.000 We have a few presentation about sync processing. 16:05.000 --> 16:08.000 Daniel, we'll be talking about fear filter design. 16:08.000 --> 16:10.000 We have another presentation over here. 16:10.000 --> 16:13.000 And then, we have a few presentation. 16:13.000 --> 16:18.000 I'm really happy that the certain colleagues accepted to join. 16:18.000 --> 16:23.000 I would like to show you that there is a fantastic open source open hardware frame, 16:23.000 --> 16:29.000 called Y-Trabbit, different familiar for synchronizing distributed computer networks. 16:29.000 --> 16:33.000 And there are ideally suited for software different radio synchronization. 16:33.000 --> 16:39.000 So, TWSDR is going to show us how, from my perspective, SDR is one big SDR, one very big SDR. 16:39.000 --> 16:45.000 And I try to complement TWSDR's work to show you that you don't need to be a certain to be participating. 16:45.000 --> 16:50.000 We have a couple hundred-two or a worth of FPGA, you can do it at home as well. 16:50.000 --> 16:56.000 Andre will be talking about machine learning, because everyone is talking about AI at the moment. 16:56.000 --> 17:03.000 And if Arist will be taking the mic again for presenting his work on the TWSDR. 17:03.000 --> 17:10.000 And I'll be concluding about a sequel to last year's presentation about the max 2771. 17:10.000 --> 17:16.000 So, the low-bit count A2D-converter dedicated to Genesis. 17:20.000 --> 17:25.000 Thank you for watching.