 What? Yeah. Oh, man. Someone else? Here? It's going to come out? This first? This. This first camera? Oh, wait. That's a cool new camera. Ha ha ha. How much do you have? You know, I kept my life books in box. And I ended up using it for years later. Yeah. It was so well designed. It was a beautiful jet in this box. Kelly, what was that? I put a computer in the box, so I bought a laptop. So when I put this session on the PENTA schedule, are you all good over there and happy over there in the corner? Are we good? Yeah. Okay, great. So I originally stuck this session on the proposed sessions system eons ago, it seems. My original thought was that it might be fun to do this sort of beginning of the week, and if folks were inspired and thought there were cool things they could see, they could go hack around and maybe actually try some software out on my hardware over here. But that obviously isn't how this all worked out. I'm just tickled that all of you thought this was a good way to start the time on the afternoon of the last day of the event. Come on in. You are. So a better thing. Thank you, sir. You're welcome. I'm murdering you. I'm just flattening everything. So in any case, there's like, I've been creating information from different sources for this general topic of software from video for quite some time. So I actually have a slide deck here that's overloaded with lots of interesting stuff. And rather than advantageous try to zoom through a slide deck and give you no time on any of it, I will instead promise to make this deck available as I do all of my others sometimes soon on one of the sites where you can download it. So what I'd like to start off with is get some sort of sense of what you're interested in and what we ought to spend the next hour and 45 or so talking about. I do want to try and close a couple of minutes earlier than the schedule suggested in the night because I'm very interested to see what happens in this lightning talk session, which I think starts at 3.45 or something like that. So I'm feeling very flexible about all this. I'm happy to hang around and play and show and tell as long as anybody wants to, but I thought that would look like an interesting session. So let's see if that works out. So there are sort of several chunks of stuff that are happier. There's neat hardware bits and pieces that enable you to play with software from my radio. As much as we might like, it's not a 100% cool software thing. There are interesting sort of political aspects of the whole software-defined radio space, in particular challenges before our national communications regulatory bodies that have a handle on the balance between making it possible for people to experiment with new technologies and new methods and the need to regulate things, particularly for public health and safety and communication. There's also the specifics of the GNU radio software and how that gets packaged for Devin and all of Devin's derivatives. And then there's just sort of goofy, fun, show-and-tell things, gale things that's fun to use. Software-fine radio includes dancing with signals off the moon and tracking amateur rocketry flights and all sorts of other things we can talk about. So first of all, how many of you have ever done any sort of digital signal processing or software-fine radio stuff? Took a class in college even there. Okay, not many. Okay, so clearly a little bit of what this is about and basic background of being a lord. Give me a sense about this. How many of you are more interested in sort of the politics of this and cool things you can do and how it fits in larger scheme of things versus the alternative, which is you actually like to do stuff and you like more nuts and bolts kind of things. And some of both. Okay, all right, I want to hop around a bunch then and I'll try and have this reasonably coherent. If it's not, if it feels like I'm missing something or something's confusing, please turn away in the fish-size room. This many people, I think we can be reasonably interactive and we shouldn't have some new problems. So, I have the stuff that I have in this slide that concludes some final concepts of software-fine radio a little bit about the Caneer Radio project because that's one of several large but perhaps the preeminent code base for actually doing stuff. And it's one that happened to personally be engaged with. In fact, I was very pleased the other day to receive finally explicit permission from my employer to do the necessary copyright assignment on paperwork with the FSF so I'll be getting committed access to the upstream hard drive to help to ensure that things that matter as we're doing the packaging trying to make all the contributed examples actually work and that sort of stuff. I'll be able to work on that more directly in the future, I hope. I've got some stuff in here that the politics of open-source software-fine radio and it's interesting that the intersection of open-source and software-defined everything and seems to generate interesting political questions and problems and I think that's worth talking about a little bit. I brought a couple pieces of hardware. It's not some of the most recent things actually but most recent ones, A, I haven't had a whole lot of time working in, B, I'm a little bigger. I don't think I've ever had been on the air for quite as well. But I'll talk about interesting hardware stuff and then I can show you some examples where you can try some demos. Sometimes they work well, sometimes they don't work well. So this is sort of the block diagram of a typical data radio consider. And when you look... Is that... Okay, thanks. That helps a lot. So again, I apologize and have a little bit clicked on the right. All you're losing here is the rest of the words data out at the top of the data handle. The basic notion is you've got the antenna open on the left-hand side. There's some filtering. Things that look like sideways triangles or amplifiers. And so in the received channel at the top, you take the signal in command and if you do a little bit of command filtering it goes through an amplifier. It's filtered again. Circle on cross and it's a mixer. That takes two frequencies in and gives you the sum and indifference of them out. And with another filter, you can pick which one of those two mixing products you want, the amplified one. There's this little thing called eMod here that eModulates the waveform so you get bits coming out of the right. On the transmit side, it's very similar. You take the end, you may do some waveform shaping. In effect, in this particular example it's being used to control. There's a little piece of what's there's a line missing on there too. Sorry. There's a piece in effect that's controlling which of the transmitter received oscillators is being used. There's a phase lock loop circuit there in the middle that is locking that reference oscillator to produce the frequencies that are being mixed with. And on the transmit side, you're basically just modulating an oscillator amplifying it not through the filters to the antenna. So this is sort of what a typical data currency that looks like these days. If you open up, you know, any kind of radio device you'll see all sorts of analog electronic components in. These filters are implemented with inductors and capacitors and resistors amplifiers are usually either transistor circuits or operational amplifiers or more recent things like the next which are the microwave model with the integrated circuits. A lot of times big chunks of this are integrated so the whole PLL and oscillator thing might be one negation from somebody. But there's a lot of complex analog circuitry out there. This is what we'd like to have and ooh, my half-ton stuff isn't showing up well. Through some simple RF front-end to an analog to digital converter at which point it's all bits and your code goes there on the transmit side code generating bits which goes through digital to analog converged a little bit of amplification filtering it out to the antenna. Now, unfortunately, nothing's ever quite that simple but the notion is get the bits as close to the antenna as possible. On the received side we're going to get from RF to data and on the transmit side we're going to get from data to RF as close to the antenna with this little analog stuff in the way that you can. So what can you do with this? Well, you can do all this sort of conventional radio stuff. One of the cool things is if you have, there's a quote down here from somebody playing a particular type of software-defined radio that's popular in the hand radio world feature sets, modulation schemes, filter performance all of those things that, you know I don't know how many of you are old if there's a point in history where when you buy a radio they actually told you how many transistors it had in it on the outside of the radio this was a 7 transistor in the radio or 13 transistor in the radio it was really exciting to have more stuff in the radio. And then there was a period where oh, this one gets four short wave bands and two medium wave bands and with software-defined radio so there's an easier A to D and D to A are they good enough and there's a little bit of analog electronics you still have to have good enough. And beyond that it really is just a software-enforcing problem. And it's possible to so it continues to happen. It's also possible to achieve performance that's just not possible in analog circuits. And if you ever took an analog electronics class you'll remember that whole section they made you plow through about filters the quality of the filter and the gain of the filter and the ripple and the passband of the filter. Well with software-defined radio techniques you can have filters that really do just look like they have square edges they have a nice flat gain through the passband and they get cut off amazing numbers of laptops in short numbers of in short businesses on either side of the interesting thing. You really can do things in software that's just not possible to do with analog or electronics. There's this other interesting thing that happens that in digital communication systems there's sort of this tryout of bandwidth time and power. And if you want to improve the quality of the link you can actually consume more bandwidth spread the signal, use coding techniques or you can transmit the same piece of information over a longer period of time which might mean going to a lower data rate it might mean repeating the data more than once as partisan forward error which means where you can find the power and what's interesting in software-defined radio techniques is that you often have an easier opportunity to make these trade-offs and to use more bandwidth and more time because it's under computer control and all the software is involved. There are all sorts of other things you can do. I think this notion of cognitive radio is a really interesting idea and it's one that lots of people talk about and it's still to be fully explored and you have a radio and it comes up on a band and there's some interesting signal there and the radio figures out what the modulation scheme is and the radio figures out that well, to talk to that person today we ought to use this kind of modulation scheme as kind of a power level and so forth. We've seen little bits of this today all of you that have JSON and mobile funds they all did power control and the way they did that is the cell site that you're talking to and you're reporting to get the quality of signal that you're receiving and there's this little protocol dance going on quietly in the background as they figure out how to power control each other so that you don't overload the central site with a stronger signal when it really needs which would make it harder for you to hear all the other handsets and so forth and making all of those things work cooperatively together well is why sometimes we end up with regulatory issues that surround us so then what's getting ready well it's a free software toolkit for building and deploying software and fine radio things it's also a really great way to learn about digital signal processing I should mention that 20 years ago everybody talked about digital signal processing well today, more likely you'll still hear about that particularly in instrumentation when it comes to communication the thing we call software and fine radio is sort of a combination of sound processing and other techniques it's kind of a super stuff if you're going to do useful stuff you do have to learn something about communication systems and you can't magically go I want to transmit it there are a bunch of part of being examples with the radio stuff that will help you get started but you kind of have to understand what a modulator is what a demodulator is what analog and digital conversion is about there are really great textbooks out there that are used in things like college level signal processing classes that can be very helpful but a lot of folks just in the same way with other open source projects they'll go look at some existing examples ask some questions, try some things experiment it but the exciting stuff is you really can create all sorts of new radios, modulations, protocols evolve in environments and just explore and do cool stuff that you might not be able to afford to have access to the technology or work with otherwise sometimes it's been a whole lot of time but the general notion is this is sort of a hybrid system where the actual signal processing guts are implemented as efficiently as possible in C++ and then there's a python interface that's exposed so you stitch together actual radios by instantiating these various modules from or objects in a C++ environment and they all get sort of tied together using this dataflow abstraction so you have this notion that I have this source of data over here and I'm connecting its output to the input of something else and its output to the input of something else and sometimes you have multiplexing or demultiplexing or splitting or combining blocks and you could do this I mean conceptually think of it visually but in practice I want to show you an example which is really pretty straightforward and the nice thing about this is there's sort of a separation between the signal processing wizardly food that has to go in to create blocks and put them into the new radio system and what it takes to actually sort of connect them together and sort of think in terms of I need a filter with these characteristics well necessarily having to understand all the GUI code that's going on or anything like that for it it is more involved in creating blocks but it isn't already there there's a very vibrant community that is happy to help you create that so here's like a little example bonus points to anybody who can tell me what this goes you're losing two arguments on the ends of the six source lines but I don't know if any of you have done a lot of round line up actually it's also sort of a US example what's going on here is really simple we're instantiating two signal sources here sources are in source one they're oscillators that are being defined in a pair of audio frequencies and they're being connected together to an audio sync do you see a left one they're pressing a button on the tone it's a thousand it's what you're hearing in stereo let me see if I can remember where I parked that there it is yeah so it really by the way just so you don't believe there's something weird going on what's the thing that should control the bus just so you have some idea of what this really is this is a couple of commands in the process but there's that same thing that I was just showing you and that's all I'm doing so it gets more interesting when you have a signal where it already is and so forth so there's this other thing that goes on that you have to mention in software to find radio a lot of times what we hear about are these sort of political issues and in particular there are problems with the fact that today a lot of wireless network interfaces and notebooks are actually in communities in software to find radio techniques and it's because they're implemented in software to find radio techniques that we get a lot of cool features we get from price points that we get but it also is the reason that it's really hard to get a open source wireless interface in a notebook these days because there are regulatory bodies in various countries around the world I keep being told but I don't know the details that Japan has a particularly significant place in this regard that really don't want end users to be able to twiddle the knobs on radio too much and so there are regulatory requirements on what interfaces people can expose if they want to be able to sell these pieces of hardware I think at least for the near term we're going to be stuck in a world where we have some binary downloadable firmware for wireless cards and we don't get full exposure of the hardware interface It's too bad because there's all kinds of cool things you can do if you can program the radio hardware that's directly in either the Wi-Fi interface or more interestingly at least right now in the US with the sort of cell phone card radio interfaces that are built in machine but you know, what we can do we can do and what we can't we can't What ends up happening here is that this isn't a simple one way thing it's not one of those freedom is good and we deserve this and we're being mean to this kind of problem it's much more complicated than that and the reason it's more complicated is that as a society we have to balance this concept of freedom to innovate against useful kinds of control I think one of the best examples is really what happens when we're talking about public safety communications I've had the experience in the past of traveling around with people who do things like provide the emergency communications for search and rescue teams up in the mountains in Colorado and it's not unusual for those guys to have a vehicle with seven or eight or nine different radio systems installed in the vehicle and these are all different agencies that have different spectrum allocations different ventures of equipment modulation schemes and all this and it's tempting to think of you know could we just have one box that you put in there and understood about these modulation schemes and frequencies and use more software and techniques and yes that would be a benefit but then you turn around and you say and what happens if there's one guy sitting in his unmarked sedan on the corner of the parking lot with a jamming transmitter that you know works really well because he's got all the same software to find techniques and he's preventing these public service guys from being able to talk to each other and go rescue somebody who's in the dark streets and so I think there are legitimate reasons for regulatory bodies to want to be able to control some aspects of this stuff and the trick is just to find what the right balances are I don't know what it is certainly through history part of the motivation or the radio project has been things like the broadcast flag that I'm sure many of you have heard about on high definition TV broadcasts the way that sort of intersects with the digital learning copyright act in the US in this notion that this is a mechanism for protecting content and if you reverse engineer if you've done something illegal and yet we're in a situation where if you follow the way that's written on it there's no provisions for fair use of those streams at all the good news is there was a court ruling in the US that said that the Federal Communications Commission only has the right to regulate the part that's going over the airwaves and through the transmission networks and once the content's received they don't have the power to regulate that of course immediately there was then a request filed by Congress to write a new law that gives them a different set of powers so this is an ongoing thing it's not a one shot deal it's not oh we won the war oh we lost but it's something where we have to successfully educate policy makers and decision makers that this is one of those places where there's an opportunity for innovation and invention and for people to experiment with the new cool new things and yes it needs to be balanced with whether the behaviors of the end of the day are in the public good or not but it has been a case in the past and I think it's still true that at least some branches the code tree that on the radio stuff can do a full software to curate the modulate and display it on that TV broadcast but frankly ever since the broadcast flag stuff progressed that's been a less exciting thing so ok so if you want to actually play around with this in that live diagram show that you have to kind of get from the RF world in and out of the computer and you need analog to visual and digital analog conversion and depending on what kinds of A to D's and A to H's you have you may also have a need for some simple pieces of RF electronics to get yourself converted up and down from frequency bands that are really of interest to you and within the sort of amateur software fine radio community which is to date largely been around the amateur radio world though there are university projects doing cool things and other people as well there are two main approaches to this one is to figure out how to do things with sound cards and in particular to in effect sort of do the modulation and demodulation part but not necessarily a whole lot of the radio part in software fine radio and then the other approach has been to have higher speed data interfaces like USB 2.0 which these ports use and to have higher speed A to D and D A converters that are dealing with frequencies to get at least up into the IR range not all the way to RF so if you're doing the sound card interface the good thing about sound cards is lots of people have them and they're cheap and the bad part in some sense is that you can really only use these with narrow and low frequency high ups it's really good what you want to do is listen into something like a paging data stream or if you want to I don't know we've talked about all sorts of interesting position reporting and locating systems so if you want to make it possible for anybody with a notebook and a scanner to be able to listen to and decode the data stream that's coming down from your high power model that you just launched they can help you to find it there are things like this that work really well in the ham radio community there are a lot of people doing things like this this is my deletion scheme called PSK31 it's actually a 31 and a half broad communications mechanism which is really really slow right but with PSK31 I have sat at my parent's beach house in the other banks of North Carolina with a short wire antenna thumb tacked to the railing of the deck a few feet off the beach a 15 or 20 watt transmitter on a notebook computer and successfully contacted people in 23 countries in 3 days including New Carolina which is more than 12,000 miles of that path and what you're gonna do is you're just sitting there on a keyboard to keyboard typing with each other think in a person to person IRC message or query kind of stuff everyone has a few characters are garbled but your brain doesn't care the person doesn't type that well so whether it's actually being lost in the RL they just can't type your brain sort of ingrates this and you can have a really good time and have nice conversations with people and there's no infrastructure involved there's no internet there's no telcos you're not paying pure IRC it's basically point to point communicating directly with other people across political, geographic national boundaries and all this and there's no infrastructure even to it's too slow for fables and it's really too slow you make a good point and certainly in certain countries you're not allowed to discuss sensitive things well and there's a there are we won't go down that level but there are the interesting thing to me about this is that there are so many people around the world who are not good at typing a lot of the programs people use from PSG31 have these big macro facilities here's a complete description of all the equipment that I'm using in my station here's a complete description of where I live and how far I am north of such and such full town you've never heard of and here's where all my kids are currently going to school and they'll sit there and sort of push a button and then it takes about three minutes or something so it's interesting that if you're in a contest situation or something people are going for the maximum number of contacts that's sort of interesting because they're really quick and short exchanges and kind of content through on some level and then there are these, they're sort of sitting around and around in the evening having these long and often broken content but then there's but then there's also the case where you run into somebody who's really interested you know like the guys talk to one evening on Monster Rock Island doing some consulting work and we got an hour and a half conversation until the art of connection we had finally sort of faded due to change in conditions in the ionosphere but this is a total loop talking zone and so these are things that are just fun but there are also more things that people have done as well so this is the guts of this name hold on this to everyone in the past and I don't actually have one of those with me it's a small board stack that implements hardware that allows you to get from a sound card interface on and off the HFM radio bands very easily and in fact yeah I have some clips a little more challenging but it basically gives you a receiver that's good up to about 65 megahertz and will transmit on the HFM radio bands and it's possible to have it with up to 100 watt or even a millimeter there are all kinds of features on this now but for those of you that only think about filters the fact that you can have a 25 hertz wide more scattered to the safe filter that doesn't ring at all it's a really crisp almost perfect filter it's kind of mind-boggling and the software features on this just keep evolving this is kind of what the box looks like it's about so it's not totally tiny but it's in effect a complete HFM transceiver it's an equivalent to the multi-thousand dollar Japanese and German made transceivers it's well under a thousand dollars and interfaces with a sound card and you can use some relaying stuff and there are people playing around this is one of the user interfaces that somebody called up as sort of look more like a communications receiver on your screen close up so then the thing I personally spend more time with is why Bandai is such a discussion that to be fun to do things a little more directly to our and where it would be fun to be able to experiment with wider bandwidth modulation schemes than people normally play with on the Asia path some of the earliest experiments that people were doing were with PCI bus plug-in cards that had A to Ds and D to As those are usually made for use in industrial automation process control, prototyping so they're very expensive and you still need an R to run out because that's just A to Ds and D to As so more recently meaning in the last three or so years I guess four years there's kind of focus on building things with USB 2.0 interfaces and it's a cheaper and more reasonable way to get into this there are now people sort of fanning out again and there's discussion about other interfaces that are becoming common on computers but personally the things I'm playing with are still mostly USB stuff so one of the things I have with me is one of these SSRP cards which is a David Cardes on this little daughter card and underneath it is underneath is a board that came from somewhere in eastern Europe actually that's just a developers evaluation board for one of the Cypress USB 2.0 microcontrollers especially USB connector or microcontroller or crystal connector and two passive components from power supply part and then this daughter card is the software find radio piece and this is a received only no transmit side at all he's trying for a while to be working on an equivalent transmit board but I haven't actually seen any of the parts being made available yet so this basically has a connector for hooking up an antenna or a feed line or something an amplifier and an A to D with some filter and an A to D convert and stick in as a crystal that sets the sampling frequency and so this can be used to get RF sampled and into the computer as data actually it's reasonably robust and I don't go gnashing it how much of that did you say? I don't remember I think the little micro processor was in the like under $50 US range and I think I just got the bare circuit boards for the daughter card from David and I think I paid him $10 or $15 each for them the A to D converter and the amplifier that are on there are routinely available at least in the US it's free samples from the mini characters if you ask them nicely and have them look something business address so we're academic in fact it's interesting because this whole free samples thing you might feel like it's a scam or something to be telling them you want this for having things but every time I've ever talked to people at those companies they're actually pretty enthusiastic about folks like us getting a free part or two and playing with them because it builds the body of knowledge and people end up actually occasionally building a product and buying tens of thousands of units as a result so that's just part of the dynamic about that I never lie to people when I'm requesting free samples if they have the thing and it says is this for a product that's about to go into production it's clearly not the box to check one down here if you know I'm thinking about it or evaluating technology that's a good box to check when they say how many units you buy in the next year I always pick whatever the lowest volume one is zero to 100 units or something I'll figure if they still want to send me the parts and they must really want me to vote there's another thing there's a lot of your fellow who I don't know very well did it's kind of an SSRP equivalent there's the USRP board which I have here that I'll talk about a little bit more that Edis did and then there's the newer high performance self-proclaimed radio project which by the way it does view with notebooks that want to go surfing ahead that's hpsdr.org there's some cool stuff there but as we haven't seen it yet this is that SSRP board this is the microcontroller board that goes with it and that's the daughter card and once again sorry for being a little clipped on the right my notebook despite Keith Peefoo on this projector we're not entirely happy with each other but there's SMA connected on the lower left that's an internet connected part a little black part just above it with eight leads it's an instrumentization amplifier it's a high quality amplifier the passive components behind that are going into that linear technology and it's basically just sampling the RF and giving the numbers out representing the amplitude of that signal on regular intervals they get passed back over USB 2 to the computer how much of that the RF specific versus just playing an analog sample or some other kind of analog sample the only thing that really ends up being RF specific is the very front end where you're getting a little bit of signal conditioning so one of the things we want with RF a lot of times is the ability to sample right down to zero hertz and DC offsets whether you're a DC or DC couple to be able to convert, things like that matter but not all of them matter as much to people who are doing pure instrumentation stuff so this works and it's kind of cool but the problem with this kind of interface is that you're generating a pretty hefty data stream as you start to go up in frequency because there's no logic on here to decimate the sample if I put an oscillator in I don't know what the one is that's in there you're going to get samples that often and the computer's got to be able to eat that data stream coming in over the bus and do something useful with it and if what you're really trying to do is you know to code the audio to figure out what the combination is to that point there are a lot of things that you might want to do that don't need a lot of bandwidth and you end up consuming a lot of computer resources that you don't necessarily need with a dumb-ish interface like this the reason this whole software fun radio thing makes so much sense though is that Moore's law and the economies of scale mean that the computers we have and the abilities that we have to come with them and because of our buddies doing gaming and graphics and so forth the cool, single processing ish instructions that are buried in common processes from Intel and AMD these days are just stunning for doing software fun and radio stuff so it's amazing how much it really can do with a generic notebook and these sorts of interfaces but at the same time this is a relatively simplistic interface, it's just a native e-converter that generates a big stream of data I really don't have problems pushing so much of the USB that was more or less my problem well this is why this ends up being so limited in the end this is Vladimir's board, it's very much like the SSRP it's got an analog device and then the big black part is the USB 2.0 microcontroller the other side of his board is pretty boring first of all few resistors and chaps and LED, some connectors not a whole lot to it in fact his header is down here mostly to get access to other pins on the processor by the way I don't know what the status of this board is but it struck me at one point that it'd be kind of cool if somebody just made a bunch of these for folks to play with we don't want to set it up when it's hand-done with chemical action out of the print PCA yeah, yeah, yeah, yeah, this guy's and I forget where he's in some interesting part of the world that doesn't have as easy access to some of the services that we did in the US at least we can upload a design file to any of the number of little companies in the US to get access to your $100 U.S. formula and deliver prototype multi-layer boards back in two days in fact you can do that for getting aluminum or steel machined if you want this little spring from panel hard work down to all kinds it's almost the resurgence of the ability to hack on hardware stuff is that a lot of these processes that are hard to do and if intimidated people even surface mount assembly and so forth there are a lot of little companies now that popped up and have a web interface and get a credit card and are passionate about something in a few days you can get things turned around and sort of, you know, it's that it's that personal version of outsourcing processes so this is kind of cool so then the USRP which is this board and this one has some dollars and sort of the next evolutionary step and this is, in fact the SSRP, David Carr did that because it was taking a while for not to get the USRP done and David wanted to play on meantime or something like that but this board is really neat because I've got some photos I'm going to capture what this has is the same USB 2.0 microcontroller a couple of dual A-to-D A-parts from analog devices and a big Hong Kong and Altera part in the middle of the board the Altera part is a dual programmable data write that's soft programmable so when you first plug this board in on a USP interface it comes up dumb and you download firmware for the microcontroller part of which is the bitstream and then it does the USP reset and re-identifies itself as a USRP with a particular firmware load and off you go to the races what's cool about this is that you can implement some of the high processing load front-end functionality within this FPGA so the amount of data that's going back and forth over the USP bus is sort of what's important for the thing that you're trying to implement and a really sweet thing for folks like me even though I've done FPGA it's never been something I've really enjoyed there are a couple of standard loads of functionality for the FPGA that do routine useful stuff that are available in the new radio so that you don't actually have to become an FPGA program wizard to take advantage of that and then the different daughter cards that are on here have different kinds of interfaces this one with the shiny aluminum is actually a UHF TV tuner that is an after connector US ridiculous RF connector I don't know if you're stuck with this chunk in this part of the world or not but you put a antenna in here this covers up to I don't know 860M or something like that and it'll down convert and then basically the output of that is going to be better than the A to D these ports are called basic RX and basic TX ports they just have connectors to isolate the RF connectors from the A to D a little bit and all these other headers are actually so that you can plug logic analyzer probes on here if you want to if you're trying to do things like debug what's happening in an FPGA or otherwise you're interested in doing more sophisticated work with and in fact if we try to play with this a little bit today it'll mostly be with the basic daughter cards because there's some cool things you can do even if you don't have any RF here how many gates can you put in that can't you get it? I don't know, I think this is a 100K gauge piece but I wouldn't spread it out I'm not an expert on the welcome pool cyclone this is the kind of stuff that some of the standard was and again I started to outline a couple of boxes and missing here but it's the rest of it's an oscillator it's implemented and everybody gets a chance to look at this or do they not get around if you guys all go okay fine and there's some decimation going this is the notion that you can run the 8 and E converter at a high sample rate but then only take every nth sample to send back to the computer because one of the things that I think people forget is even if you have a signal processing class and you learn about the Nyquist rule that says that the bandwidth of the thing you're sampling is at the sample rate that's bandwidth not frequency but a lot of times it's useful to have a system that will hide a sample at a higher sampling rate and be able to in fact carve out just a small chunk of bandwidth that you're actually going to work with so to minimize the data rate and the decimation function is what is that and then on the transient side you have a similar sort of interpolation feature that solves the same problem so here's what one looks like with a couple of the basic and then before we get away from talking about hardware stuff I'm going to talk a little bit about the HRPSDR stuff this is an interesting project to me for several reasons not just the fact that it's offered a fine radio one is it's one of the first pure open hardware projects I've seen in a while and what I mean by that is these guys have gone so far as to motivate an intellectual property lawyer who is former president of Tucson Amateur Packet Radio pretty well known in this community to create for them an open hardware license and so there is now a thing called the OHL an open hardware license and you can think of it as being a very GPL like for hardware designs and this is something because this community observed the neat effect the network effect that happens in the open source world when people agree to share and enjoy and collaborate and they wanted to do the same thing with cool hardware designs and hardware is just a half different from software in terms of its characteristics and what you need to protect in order to have a meaningful copy left-ish behavior downstream that they needed a new license if you're a native and hobby hardware hacking and don't work for a company that has intellectual property concerns and being able to do this I would encourage you to look into the OHL and think about its possibility the other thing is these guys are very much again interested in creating building blocks that other people can use to do cool things so the notion here is it's a small volunteer project creating high-tech modules for experimentation and advancing the radio art the designs are open source hardware software all the programable logic and so the translation is a bunch of keys and I don't think I need to spend a lot of time on it here but this notion of open source hardware I want to call it that there's been something a lot of folks in this community have had to spend a moment to wrap their brains around this notion that any changes you make to somebody's design and creating a derivative circuit board for example are things that you should be able to share or you should have to share with people downstream and provide fixes back to the folks that you got it from the notion that the design should be freely available when one asks and there's it's just it's an interesting concept when you start thinking about this and what it really comes down to is it's not clear that companies participating in the production of commercial hardware would like this kind of a license at all it's probably even tougher to contemplate a major corporation directly engaging with another hardware project than it is today to see major corporations and software patent portfolios participating in open source projects but who knows the world may get us there at some point but what is cool is it says that all these folks that are doing things on their hobby time and learning and from each other in the process are building a body of work that is also shared in the use of new tools stuff so this is actually a backplane based design for those of you who have seen Eurocarts before will for BME best stuff will recognize their choice of background connector and the spacing of all this design so that this could fit in something like an AT PC chassis or an ATX chassis and the card to bottom of the board and it uses an ATX style power supply this particular board is being powered upon as well daughter cards that were made for ITX boards and so forth to run them off of DC it's an interesting thought there are a couple of boards that are initially available that are interesting Ozzy is a board that in fact provides the equivalent of the logic guts of the USRP and that it's principal purpose is to get into a similar newer generation of the Cyclone part the general notion here is a way to come in from some other computer using USB2 and get onto that backplane and have some of the FPGA kind of functionality and then JANUS is the in fact the A to A and D to A converter piece of this and the reason that this is the first thing that focused on and the reason this has been so important is that I mentioned earlier that one of the nice things about the software from radio stuff, the sound cards is there's so many of them available the problem is that so many of them share the problem and if what you want to do is play with PSK31 while you're featuring something it's okay that's not super demanding signal processing application but if you want to do something like weak signal contesting in the amateur radio microwave world or something like that you really need a better quality radio and in sound card interface software from radio stuff the quality of the radio and its performance is directly behind the quality of the sound card how linear the A to D and D to A behaviors are, how much from the ways there is and so forth and when you go look at some typical commercial sound cards using codecs that are so typically used you see these interesting effects you notice how towards the edges of the spectrum they're going relatively normally along the rights just for once you get out very far if you try to do high performance software from radio stuff you can compensate for a little bit of this but at the end of the day this is the thing that lets the signal to noise ratio of the system therefore fundamentally the performance is constrained by the quality of the sound card this is what's possible if you pick the right A to D parts to build your sound card and you get the signal processing stuff right so just a quick comparison okay time for a little spike in the center versus B to A something in the center no funny little side spikes in the center versus bunches of them a nice flat way out to infinity versus you know what the heck is that over in the right it's like a gap to occur or something well it's kind of okay for a sound card to be a lot of the quality of the parts and totally pass that because after all the humidity here has certain intrinsic limitations and you know for talking about sound cards it's so good isn't it like when you're sending files what is the signal you're putting out to the sound card in those graphs well I'm trying to find out what are the graphs actually showing inside the frequency so this is frequency across I'm pretty sure this is just noise level okay wait is it a single tone the spike in the middle okay so you're sending a single tone I think that's correct so then there are other things that people are working on so those are the parts that are actually available today is this basically a three board stack that's available it's the backplane the usb interface with the big pga and the thing that looks like the sound card on the other side and that three board set gives you the equivalent of a usb attached really high quality sound card for doing stuff up to the 100 to 100 hertz ish kind of range which isn't bad because with something like sdr-1000 providing a lot of hardware interface to convert up and down to various ham radio bands the combination of this board set and something like the sdr-1000 and your average ham radio user software found radio is going to be pretty happy for a while but then there are other things going on this mercury board is a really fast beta d board that's going to have performance capabilities slightly better than even a srp and it's current version yeah this is another interesting board I don't know if this one is still live or not or not you very quickly get in this mode where there are some things that are clearly building blocks intended to be put together on that backplane and some of them are projects where the board might end up being something that could plug into the backplane but they're designed to actually be sort of separable devices so in this one you notice there's a canvas interface down at the bottom that's because this might actually come out of a satellite using one of our can control interfaces and you notice it's got a big dsp chip in the middle of it so they can run standalone without needing to have a processor out there so some of these combinations get kind of interesting but also within the high performance srp community there's a lot of interest in precise timing if you want to do software found radio things over long distances with complex modulation schemes knowing that both ends have their clocks locked within some number of nanoseconds of national reference allows you to make assumptions in your modulation schemes and protocols that you couldn't otherwise make some others just like and precise clocks so there's all sorts of interesting stuff happening here with precise timing interfaces that would be part of the system as well would that be nearly for a frequency update? well that's one possibility but you can also imagine for some of the moon bounce things that I'm interested in what you'd like to be able to do is get really long integration times where you're transmitting a coded signal frequency and in order to be able to accumulate energy over a long period of time from that signal you need to keep the frequency that you're listening to really steady and you need to know when to be listening to differentiate between when the other guy's transmitting and you should be integrated and when he's not transmitting then it would just be noise and the bucket we actually have protocols now for doing things like sending short messages via the moon where the coding is super intense and it's possible in software to detect these signals and have communications when the signals are 20 or 30 pp below the noise so then you're just doing a scurred event noise averaging examples to increase the and also doing minimum coding tricks so that the minimum bits can go in tricks so that you figure out what set of messages you might want to be able to exchange and you can design a coding system that sort of uses a fixed interval of time and uses the maximum coding food to transmit a really short message with a lot of repetition so that you can integrate at the time and all kinds of tricks like that and a lot of those depend on having real-estate loss letters and it's prototype expansion for what was the simple thing. Because it's hard. Also you can actually from both sides of the moon you can actually talk to someone on this sort of planet. It's actually one of the little niches of the anti-radial habit that I find intriguing and so I had a three meter dish that I had retrofitted and built an ASL mount for in my backyard that was dedicated to doing this kind of weak signal stuff and it's total good just sort of going out there and lying the dish upon them and getting to the right place and hearing people that you know in Switzerland and Italy and here in the UK and each other off. I'm kind of proud of you. When I get to the point of actually participating in the talking half of this and not just the listening part of doing 500 watts of 1300 megahertz 500 watts into a three meter dish in the UK so don't stand in front of me. I have the equivalent of make about 300 watts at 2.4 megahertz and about 130 watts at 3.5 gigahertz and about 75 watts apart from 7 gigahertz apart from that. We've got to do that in front of my laptop. Do you have to tell me what's falling down or what's that? You know if you do the math on this the energy density is actually very low even a few meters in front of the dish it's not a problem it's not like I'm going to be roasting a squaw or anything. It also helps with a little bit of high altitude as far as what that's another story. Now people routinely do this stuff and in fact on the backside of the dish it's just kind of a reflective surface and so the zone that you need to be careful about is fairly close to the dish and so I know you wouldn't want somebody standing in front of it but I'll show you a photo of it it's actually hard to stand in front of a white side of it So anyway, I think this high performance SDI project is really neat and I think it's got all the cool characteristics of an open source project and so I encourage all of you to take a look and see what's going on I don't know if you guys heard there's this thing called SITSAT where the amateur satellite community batteries into a a leader is about to be thrown away Russian space suit and they got tossed out of the International Space Station so I slowly have it to work with to generate burn up in the atmosphere and the intention was for that to sit there transmitting for a couple of weeks there were a couple of problems with the RF stuff in the end it was harder to hear than it should have been but it's still generated an amazing amount of attention and enthusiasm and all sorts of features and all sorts of people all over the world so we're going to do one of these again and we're going to put our quality software from radio technology in there what we call SITSAT too and you can chuckle when you see that community out there Okay, so the other thing that I spend time playing with that I'll spend just a few minutes telling me about and then we'll quit and kind of playing with some hardware nutrients as an example because the amount of radio satellites is obvious and again this in the context is a thing called AMSAT there is a German project called Phase 3D and an American project called EGLE that is sharing much of technology blocks and I've been helping to contribute to several of those but fundamentally a lot of neat things come together for this so you'll notice that there's some similarity to the vision for this next generation of amateur satellites and this vision I've talked about when I first started with what we'd like to do with SDR to display radios Old spacecraft, one of your transponders had this big analog IF signal condition on the block in the middle and there's all sorts of problems with this, not the least of which is that DC power is really hard to get to work with it costs a lot of money for high efficiency solar cells to carry them all with this and yet we've done some pretty neat things in the past but you end up with these fairly complex looking analog circuits with lots of shielding this was for its gear back in the day or sometime and this was an astounding piece of work but there are a lot of techniques that we can bring from digital signal processing to figure on this and SDR stuff is that we actually intend to build and fly some software-defined radio transponders from our next generation of amateur satellites Do you plan to get in and out of it? Yes, we're all fully dynamically software configurable from the ground so what's cool about this thing is every year there's this incredibly heated gathering of natural radio operators in Dayton, Ohio and then some of that but one of the problems with trying to demonstrate anything in Dayton is there are so many people demonstrating so many radio things that it's incredibly ambitious to harsh RF environment and one of the problems that we've had in the past with amateur satellites is we've had great systems until the military brought up a pulse radar in the same band and we were within a range of one-year radar sites or some guy that's got more money than the rest of the photographs down to a satellite community puts up a big honking antenna with a big amplifier and doesn't want to turn the game down when he doesn't need it and all of a sudden nobody can hear anybody else so we're going to have an interesting little experiment we took the prototype to one of these modern transponders to Dayton and sat it on the table you know, some plastic wrap over it just to in fact keep people from poking at the board and ran it all weekend at that Dayton and people walked around the event talking to each other and listening to the beacon from around the event and it's just amazing the dynamic range is more than 85 dB and everybody commented on how nice the audio was through it and all of that and it didn't get crunched the whole weekend so it was one of those it's a stupid way to measure things which is sort of not able to speak about it numerically it's a typical standpoint everybody that had a chance to play with that demo kind of went wow, this is kind of cool part of what we're doing is we find really strong signals that are not of interest in the past kind of watched them out so they're not consuming all this kind of stuff can be done with sufficient processing if it's all in the digital domain it's completely and utterly impossible to implement that kind of behavior in a purely analog system it isn't simple I mean there's a lot of parts on that board this is a black diagram of it at the end of the day the notion is that we can actually deliver communications capabilities that were never possible before by applying software and radio techniques to these signals I'm not going to worry about it but if you want to read up on that stuff, feel free so you asked about the E&E stuff yes, this is all descriptive, but here's what the dish looks like sitting up in my backyard as long as you met my son that's weighty, which is a little over now but not a whole lot that's a three meter dish as I'll mount the adults so you can scratch the sitting up in my backyard and with 100 watts and a software-defined radio transmitter I can use this to set up the moon I'm at the point now where I'm successfully hearing other stations and decoding them entirely using a software-defined radio stack all the time the modulating bit streams for the folks that are using digital modes or getting the audio of the CW for folks that are using Morse code and having no trouble with that at all I'm hoping sometime this fall to get the transmit side all put together I have to stay home long enough in one stretch and not be diverted from my semi-capital of my rockets all the time to actually get around to finishing this up and there's a contest that happens every fall that I'm absolutely bound to determine to be on the air for fully operational as well so let's see how that goes oh, you got to have my rate on that best use of a plastic garbage bag because it's so hard I'm also exemplified with some pieces of little shit that I do stuff from anyway if you want more information and want to follow up and learn more about this these are some URLs again I'll put the stack out where people can find it and then it's like 3 o'clock now so if any of you have sort of had enough that's fine I do have harder pieces here and I would be happy to start plugging things in and we'll see if we can try and play around with a few things so find a not-too-giddy-ass up-and-station to listen to with a native e-converter and a wire antenna, that sort of thing or anybody can find that interesting otherwise I'll be happy to take questions, open this up for more discussion and thank you for your kind attention and all that kind of stuff First of all politics before getting to hacking one of the neat things the S&C has been in the United States has made certain bans available for unlicensed use so for example 5 years part 15 and of course usually the requirement is that there are power ceilings on this kind of use and it seems like there's a lot of discomfort around that but I think that having unlicensed spectrum is really a great way for the open source community to experiment with things like as you mentioned the radio and more sophisticated kinds of radio do you know of anything going on at the global sort of political scale I'm thinking now a war to open up more global unlicensed bands not really it's not that people are negative about the concept it's just that there's this perception that spectrum is incredibly valuable or resourced and it's not completely renewable it's renewable in distance to everyone's use and it's renewable if you're using modulation scenes with each other and so forth but it is sort of a finite resource and there are large economic powers to work here having said that in the US which is where I know the regulations the most in addition to part 15 unlicensed there's a part 5 experimental and there you have to apply for and receive an experimental license to do essentially anything on any band you just have to make a case for why this is something it would be reasonable for you to experiment with and that you're not going to cause unwanted amounts of interference to other occupants of that channel it's a process you can work through and I know people that are doing interesting projects where they're operating in the report by a license then there's the amateur radio service which in the US is the part 97 tools and that frankly it's gotten to where nowadays in most countries have relatively straight forward processes requiring hand radio licenses and if you actually want to transmit and receive experiment modulation schemes it's not hard to get hand radio licenses anymore it's still tough in places like India where the police investigation of your background is tougher than the technical license I mention this because our good friend and fellow D.D. Ramakrishnan and Luther Krishnan who was maintaining the new radio packages before me went through the multi-year process in his now licensed hand radio operator and it was well over a year from the time he completed the test process for the license until it was issued and even then I think it required some conversations in various directions with interesting people because they just I guess it's a country with a different set of thoughts and beliefs about what individuals ought to be able to do and what the restrictions are and so forth but in the US and UK and Germany and actually Australia still has a more expensive licensing process in some places because well it's not in the SEPT treaties they haven't actually implemented the rules gene that's required to put them into effect in the hand radio service test test but as a US hand radio operator I can transmit without any funny licensing pretty much anywhere in Europe and pretty much well New Zealand and Japan lots of other places without an issue so I actually encourage people to go get a hand radio license if they want to because not because there's 100% correlation just because it's an easy way to get permission to get access to lots of different kinds of spectrum and the US restrictions on hand radio are pretty palatable except for the desire of not having encrypted strings but no I don't know of any sort of global level of issues to try and create a lot of license spectrum though I think it's going to happen if there's just too many use of strings I think about Wi-Fi Wi-Fi is a good example Wi-Fi stuff is operating on license you don't have to go get a license from your national regulatory body in order to associate with an access point of error thank you Roger but this is not a use of phone eater that's because people who made the phone license that's right because you don't in that case you don't actually have specific spectrum rights you have service rights with a service provider who's licensed that spectrum this is where the regulatory gets interested because there's actually a rule the eCPA the electronic communications privacy act in the US that makes it illegal to not forget exactly what the terms are you're not supposed to listen in the mobile phone signals basically you're certainly not supposed to record and share it with anybody else never mind that everyone's lost and the politician gets thoroughly embarrassed that gets technically harder as they move towards more coded modulation schemes and less just long RF that's right that's right that's right but my point is just that with software if I'm ready to talk to you there actually was an unreasonable and yet serious proposal in the US Congress that A&E can be licensed you know this didn't go too far but it just it sort of points to the mindset of a you shouldn't have any rights unless they're explicitly given to you kind of behavior which is so counter to our historical government bias of those rights reserved to the states and adhering to the the citizens that I'm specifically talking about so it's just this is where it gets tough she assumes an encrypted body or if you're being so sure it's legislatively encrypted I've never actually seen an attack on A&E I've heard rumors of yes somebody had reportedly rolled but nobody's actually been the practical issue is just that you cannot for example if you buy a general coverage receiver it stays it's administratively locked so that you can't receive the chunks of spectrum that are allocated and they all have a warning that it's illegal to modify it so if you happen to find the jumper that you need to remove or the diode that you need to add or get that you need to flip to turn that feature off you may in fact be violating the law so these are the places where you know you want to support website yeah in general in general it's not like I care about listening to the people's cell phones it's just maybe that 865M it's a good IF for what it is that I'm building right now having a receiver that has a frequency range block is just frustrating so that's the kind of stuff where I think this is where I gave my first software time radio talk it was at a Linux conference in Australia in Amber and it was in the embedded Linux mini-con and I actually got asked to repeat it on the last of the conference and one of the repeated talks and that was cool because at the end of the talk this guy came walking up to me in Australia and he told me how much he appreciated the talk because they've been talking about software for a year and more and he never really understood what he was talking about and I was talking less about hardware pieces then more about how does this stuff work and what are the politics and how do you balance innovation and regulation and so forth and he was all enthusiastic and I get it now now I can go back and have a useful conversation about this and then the UK and EUG folks I think it was a time I was in at SONZ so I don't remember who it was but again there were a couple of tracks that came up to me afterwards and they were similarly from the national radio plays I don't know why they ended up at the EUG that they had maybe they just had an interesting unit so I feel like if I keep doing this and all of you go out and do fun things and sort of show people that they are positive, useful, productive innovative, technical things that we can do that counterbalance the occasional case of somebody who does something clearly reprehensible and gives us all a bad name as technologists but that's the kind of thing that has to happen for us to have a nice balanced picture or something and it would be really great as if we could create examples or prototypes ideally a non-licensed spectrum to demonstrate how cognitive reading could actually make much better use of this spectrum than conventional conventional coding approaches Yeah, if you go out and just do some web searches on a theory extent you'll find that there are almost anything you'd like to think of in this room today and look into the practice there may be a little one over there but maybe not English because category and other things to talk about are people to talk to a little show of interest or enthusiasm and it's often all it takes to give people and do another burst of work and get something released and make something happen and you know, look these hardware platforms are not super cheap I think this board stack area probably represents probably 700 bucks a year so it's not really cheap stuff on the other hand it's not super expensive either and it's when I think about people that I know in the amateur radio world there are those who really don't have any money and are thrilled to get their hands on a basic piece of radio gear and they'll just use the hack out and then there are people like me and you know if it's the thing that's become your hubby passion to scrape them up somebody here and there to go buy some bits and pieces to pull it over even though still got what you think an aggregate of that less than 100 bucks and again you can score some free samples and learn how to do a little soldering and things like that if you want to on this alternative we'll just start playing around doing interesting tone decoding or encoding things there's a lot of learning you can do without you can spend a lot of money but the thing I think is cool is for really modest amounts of money in a hobby budget kind of sense it's possible to get part of the platforms now that let you get really sounding stuff with open source so isn't there a tuna fish a PSK it was like a PSK31 trancey with a tuna can there are lots of things like that there's a I don't know what it was called there's several tranceyvers that are like dedicated single band and radio tranceyvers there's one that's designed specifically for PSK31 runs on 20 meters makes body walks or something sort of reports about this big and it works out just with sound cables to the sound interface and put an antenna on the other side so the only problem I have with it is I really want a fairly decent antenna if you're only going around 5 watts about 23 countries in 3 days thing I did at the beach house was I had about 20 watts or 25 watts and I could probably have done about the same thing with 5 watts it's not that huge a signal level difference but there's sort of a threshold effect in there somewhere where 5 watts is a bit of a challenge on the HF doing PSK31 and 25 watts is a pretty useful these are the kind of things that a lot of web servants see what people are actually having fun with because the one thing you don't want to do is get too ambitious and have the first thing to play with be something that's really hard I do this all the time deciding to do a moonbound station and just committing to myself that the 100% STR will be fully frequency and synchronized through GPS the national standard 100% open source for years to the project given the amount of time I had to spend between business trips and conferences and all that kind of stuff so but on the other hand it's this project for me has been cool it's a welding project it's a mechanical design and engineering project it's an RF project it's an SDR project it's an open source project my company thinks it's cool to send me around so this is about it so you know at the end of the day it's possible I tell my kids it's kind of cool when you can find something to work on that you kind of get credit for in different ways what is the status we're doing something like open GPS or something like that is it possible and what how high are the hardware to use that bandwidth or the encoding technique I actually know a lot about GPS satellite thing I helped to design a GPS receiver GPS is interesting because it's basically time coded information being transmitted in spread spectrum the RF part is not all that hard there is an open GPS project being run as part of the Portland State University aerospace thing it's a pns.pnex.edu I think see also Keith Packard out of that crowd so he can do pointers if that's not quite the right one and what they were doing was an open source implementation of the firmware for a particular flavor of GPS receiver that has a chip set it used to be GEC plus and it's now been sold 2 or 3 times I don't know who actually makes the parts now but when you're trying to recover GPS signals there's a huge amount of correlation that has to happen to recover the spread spectrum signals and what they basically make is a chip with a bunch of correlators and an arm cord on the back end on the rest of the blue closing stuff and signal processing and they have an apparently at least partially functioning GPS implementation that's GPL then but that's the only one pure software that's the only one I know of in terms of an absolutely pure GPR and I'll do the rest I haven't actually seen anybody do that I know that there's been some discussion on the new radio mailing list by people who are asking about GPS, I don't frankly I'm a little behind on this I don't know if there's been any recent discussion but last thing I recall someone people asking questions I saw someone on my university writing a master thesis on software GPS I don't know what came out of it it's technically not one of the hardest problems in the world in terms of what it would take to write the code it's pretty computationally intense if you don't have front end hardware correlation support so in something like the USRP I could see trying to use some of the FPGA implement some of that and at that point you have to ask the question well okay if I'm going to put a hardware assistant then why wouldn't I just buy a hardware chip and this is this is sort of I think the current state of things but it's an interesting thing lots of people ask about what I find and there is a working example of the new radio based 802.11 the interface but the last time I heard about it it was missing a few fundamental things like power control and the transmitter so if you actually brought this implementation up and associated with an access point everyone else would hate you you know in the access point and this is actually a really good example of the challenges faced by trying to balance freedom to innovate with sort of social responsibility but there are things like that so some of the other things that this could be some other application ideas Eric Blossom who is one of the founders of the new radio perhaps deep matter and is certainly one of the more significant members of the network or the maintainer has talked about doing passive radar where he lives here in Nevada he has a house that's sort of on a hill in one direction and there's a slightly higher and though that's a bunch of high power TV and radio transmitters on it in the other direction there are interesting Air Force facilities and he has postulated that we have a couple of antennas one sort of aimed at the stuff behind you that you could look for reflections which in effect could get the signature of the transmitted RF from some existing well-known licensed transmitter and then look for echoes of it coming from other directions and use that to build like a passive radar I don't think that he's ever done that or not it's not something personally all that well enthusiastic about yet it's an interesting idea that's the kind of thing that you could experiment with there are other people who are very interested in doing diversity reception things taking multiple ATD channels looking up to different antennas and preamps so one good example we're transmitting we're trying to bounce signals off the moon we have these incredibly it's the moon bounced communities equivalent of a big long building and playing war it's a discussion about what polarization to use in the antennas because when you have an antenna most of the antennas that you'd used before have a linear polarization in other words if it's like a vertical lip the signal is departing with a vertical polarization if you see television antennas in the U.S. they're all horizontal because they're horizontal to polarized signals the problem is that when you bounce a signal off of something else polarization gets reversed particularly so what we have done is a lot of times we do circular polarization where in fact you're better rotating wavefront and if you take that like a right hand circular wavefront it bounces off a reflector it comes back as you left hand it so it's an interesting concept to say okay well if you're dealing with something like bouncing a signal off the moon then okay it's going to be reversed coming back that means you could be simultaneously transmitting left hand and listening right hand or something and you're doing this without having to interfere with each other so much right the problem of course is that it's not a perfect reflector in fact the moons more of a refractor than less of a reflector so what you get back is not a nice clean reversal of the polarization you get a messy quasi reversal of polarization and if you're over there on the Earth's surface and I'm over here and the moon's up there it's not like we're getting what we actually get is some sort of you know elliptical and mushy well it turns out mathematically if you have any orthogonal polarizations like if you have two antennas one horizontal and one vertical or one left hand and one right hand any orthogonal combination you can get 100% of the energy that's coming back and so you know immediately the thought is oh gee well maybe I should redesign that feed so that I can actually have really high V3 amps both the left hand and right hand circular side so that when I'm trying to receive I can use two inputs on the USRP and be carling the signals from the two and getting that extra signal boost of actually recovering all of the energy from them back it's things like that that you can play with with this kind of stuff that you know trying to contemplate how one would build an analog radio to do diversely reception where you end up playing different plays and paths all kinds of things it's just at least for me as a software oriented kind of guy it seems like I'm more attracted to the problem I think so Has someone played using instead of FPGAs like an Playstation with itself processor which would be great to look for signal processing there's a lot of enthusiasm about signal processors for signal processing and I know people that are interested in like startups and so I haven't actually seen a whole lot of people playing with like the Playstation trying to do SDR stuff but I've heard people talking about it I haven't actually seen any code one of the problems right now though is that that architecture is sufficiently different from things we've had before it's not clear that you can for example just compile the new radio stack and be doing something useful the same problem we ran into is that there will be a processor some other things you can do really incredible stuff and they can go screen really fast but you don't do it by hitting the existing code just installing it and expecting it to all work so I don't know where in the process people are without it there's certainly a lot of interest in it seems like a good way to do it any other questions or should I kind of answer them I don't really know much about radio but my typical I like to be on a yacht on a remote island and I'd like internet to work is there anything I could be exploring in this sort of area I mean I don't need like you know I don't know if I need that excuse me, I don't see for it yeah you have me here until 11 if you were in a fixed place ready to do it is if you get an amateur radio license you want to leave from this country there's some overlap between the amateur radio bands and the active 11B bands so on a few channels you'll have to cut a bit under 5 million watts into a very high gain antenna and you're not going to have to transmit 1 watt into a nothing better antenna that's just going to hold the internet you're a tree so if you want to do a very long point then you can do all off the shelf stuff so I think it is it would be a highlight if you're in a yacht I guess there are actually people playing World of Warcraft over modems over satellite problems that's a little too addictive especially with this it's quite a big time satellite tracking on a satellite like is probably the easiest way to get on a yacht but that's like a decrease it's just imagine the lag you could also buy HDF which you just keep in an IRC-3 series face and then you can keep your feet out of that and that's the best way however keep in mind your knowledge is in here yeah that's very good yes keep everyone to IRC that's the responsible for content I feel like a poor boy it's what a club calls for good luck I think we're going to set up a science session so traffic is going on at 44 what do you guys think what do you guys think what do you guys think what do you guys think in a mode where it's not encrypted but you are much more full integrity can't actually you could run commercial s3 s1 in that mode because there's never been an option to run commercial s3 s1 in that mode s2 in that mode doesn't exist what do you guys think what do you guys think what do you guys think what does the carers about the physical you'd be surprised in that community I thought that you the administration didn't build the encryption key first look I'm not sure I'm pretty sure it doesn't have the software but it's one of the something I should express what is this the best we have it's just about the S3 what Colin made it more secure upstream actually I was wondering if this is something that's something that's registered things like DCI USB vendor ID and things like that it's actually a really good idea in this space to help you that sounds like a good idea I don't know we'd have to find a board member and have a conversation about it so sounds good to me we're not that expensive I don't know they store a fair amount of money for vendor ID and I think they don't need to buy a new thing in this case I don't need to look into and you know the hack that's being used here is not necessarily a good problem this is the problem you know you got like without the firmware version right you run like a usrp or er thing when I gave this package you did that with math you know plug it in well it's not really a problem because I've got I want slow run I mean this is why I'm confused the problem used to be that I wrote the I've got a fix and it's obviously working yeah yeah so I'm like what is the message it's sure this is like a read match I don't care do you remember the very girls you don't have to think about it yeah yeah yeah shiny video yeah shiny video very shiny the boxes oh yeah so call it the new one the new one the new one it's starting yeah so I'm definitely going to want a backport when someone's next so I'm doing the one thing I've got a lead then what something that's wanting for a brand It's sort of separate. It's already a couple of regions and none of this. We're going to be trying to look at it. It should be down to 60, so we're just going to say, OK. Are you going to look at it? I'm curious. We haven't only had an intracting. We haven't had a separate version from behind for a while. Oh, I just laughed. That's all I know. Maybe there's a reason I want to. There was a, I don't know, one more car. I was waiting for someone. Is there a reason? There's a stage for that. I know that we're going to try that to maintain our own necessities. Oh, is it now? We're going to try it. We have to do that. Is it now or no? We're going to try it. We've got a second one. Yeah, but you have to use some. Yeah, but you have to use some. I guess we want that one. That's fine. What's USB? What's USB? It depends. Sometimes USB is now used. There will be USB. Yeah, I mean, look, it shows up when we're in USB. Wouldn't it? Yeah, it's not showing up as the generic microcontroller. So it's done a firmware download, and things will see USB. So now it's going to be, it's a device in a case problem. Yes? Yes. Yeah. As we can use our TVR. I think it's USB or? Yeah, it's USB. So you have to write it to the right and then it's USB. It's USB. Oh, so it's USB. Yeah, it's USB. Yeah, it's USB. Oh, you didn't need it. Yeah, I'm not going to. Oh, right. No. What? It's a device. It's a device. It's just, it's just found the right device there. The program. The IOPL. They're a careless program. Actually, it's entirely possible. Oh, yep, sorry. It's a heavy problem. Actually, well, the only thing I can say, I haven't actually tried this on a 261. I'm using a 263 still, because it still has the keys, and serial things. Let's go back and see if that actually works. It's supposed to be disappointing. What do you think I need to do? Yeah, demos are down. Sorry guys. I don't remember the name of it last night, but this was on the schedule today, and I was like, oh, yeah. I think I'm going to smile or something. Can I see the bigger one? Yeah, it's the one with the plug. People at the radio would only like the key span. You used some. What? No, that one. Yeah. I used maintenance. I just want to, I just want to make sure there's a solution for the keys. I just want to show the one. Does anybody know what user stress is? I'm sorry. I know what just went on. It just got clipped. Yeah. Don't, I just have keyspan stuff built into a couple of bands that test the options. What was the, what was the one that you showed before? I looked it up and it was like, SDR 1000. Yeah, we looked it up and it was like, 2000 bucks. If you walk in front of it and buy the whole smear, yes, but, you won't want to go to the back door and steal it. Don't do that, but my point is that's with the 100 riding app and with a bunch of other stuff. I was thinking at some point how difficult it would be and the customer thought about it. There are a lot of Wi-Fi cards that have injection. So would it be possible to like create the driver that would handle the Wi-Fi stuff manually and like span several interfaces and associate to more than one access point on the same channel. Do you know something about this? Yuri, do you want to do like mesh networking or something? Yeah. Can you just set up an ad hoc mode to do that? Oh well, like, mesh network with access point, like, to be, like, I could have roaming doing pretty good, so I sent through the the most powerful access point, but without the delay of implementing access points. Woo! Well, just, you know, let's say 92 on something. That's one 25 minutes for the screwdriver. 237? Well, yeah. Anything around 96 is going to be a little bad, just for daily. Is the front-end written in Wix? What's that? Is the front-end written in Wix? Yeah. There's been quite a bit of work in the last couple of years. How long have you been communicating with it? A little bit. During everything possible, it was as bad as possible. This is, of course, the all the expensive videos that I could put on a piece of bar. That's pretty fun. What's that? I think it's the best thing that you want to do. It's a good thing, of course. Normally, you get something that you don't really need. Okay. It's a pleasure. Probably if it works. It's really cool, too. Oh, yeah. What's that? Is that a big one, too? Yeah, they are. Pretty intense. Sure. I think we need to work on that one, too. Let's get it over. Oh, yeah. Oh, I know what that problem is. It sees the the wrong one. I bet you it prefers the traditional one. Yeah. What's an HF radio? Well, don't buy software. No, no. I'm looking at the IC7. What's it? The IC4 70 pounds. It's about 550 bucks. Oh, is it like the RTL cookie? Oh, no. Oh, shit. It's... I can find it. Oh, the one I normally use is the one I use. Yeah. And they went for them February. Yeah. It's fine. It's fine. Oh, really? Yeah. Oh, shit. Oh, shit. Oh, shit. So, was that after actually just after we suspended their laptops? No. It's not working after you have the internet to the... You know what you're saying is that you have a desktop... It's not as good as somebody's for all of us non-transistors. It's something you can actually reasonably robustly demonstrate a few months of closing. Oh, yeah. There's so much more. Yeah, ignore that. I'll go, thanks. What's your frequency range? Is it really getting... Do you guys see that? Yeah. Oh, boy. Right. Come here. What's the frequency range that you can distance with this radio? It's... Maybe the converter has about a 65-meter example. So you can listen to a 16-meter or 30-meter but that doesn't have to be zero because you can sub-sample and actually specifically listen to the alias of your language. So, I don't know. I mean, the bandwidth on the basic boards is several hundred meters. I don't have any trouble with the antenna on the board. So, it's this interesting situation where you have to think both about the bandwidth of the signal processing as long as it could be an e-converter and also about the technical way of the e-converter and to explicitly think about the alias and if it's going to go on if you don't have an e-converter most real radios, it's a limited filter so you're not going to counter with all of the 10-minutes or most of those frequency markers. And that's why I'm a professor and I'm talking about the dynamic that you're going to create and how many of you are following. There are so many strange mix of the processing and I think it's done already once where you can get Jason through if they were running so that they don't have to all do the processing. People show them our brotherhood in the Nyquist frequency just to make sure that there's no various things. Well, you have to paddle through this in one way or another. One way to do it is to keep it on. Yes. I've got a bunch of interesting things that people are doing. There's a bunch of interfaces around now that have been used in its TV It turns out that those are pretty good. Yeah, it's a pretty expensive PCI board. Hang on. Cool. Oh, you're welcome. Sure. Can you explain what it was you were doing in Dayton? So it's basically a CBER a signal processing section that's regenerating the signals and varying them to be modulated for the downlink. So basically a receiver, a transmitter and a stuff like that. But we do stuff that you wouldn't think of immediately like where this problem where our downlink power is limited. And we want really neat signals coming out of the uplink to be repeated in a way that could occur on the downlink. So the system, the guy with the most power in the uplink would get the line share in the downlink but that's the way most commercial satellites work. The problem, again, is what happens if you have the guy who's got a five-lot and the guy who's got a five-lot. And so we actually are doing things like dynamically adjusting the pass-man gains of the signal that's now in this portion of the frequency spectrum that we're receiving and it gets more so we can decouple the amount of energy when the h-downlink signal comes out of the signal by shaping a sort of like taking a geometric mean or something. So basically because we have a beacon and the beacon operates at the reference level and any signal in the uplink would generate a signal that's downlinked louder than the beacon that's clicked back or someone is really slamming it or if it's a greater impulse or something that's just driving the markers and notches it because it just takes it out of the pass-man entirely so this is a wonderful tactical approach to a social material like that. You were saying that yourself. I don't think it's been in the past but a simpler version of this was that if you were too loud you would have been probably something like