Software-Defined Radio: the Apple I of Broadcast?

benfrog writes "A company called Per Vices has introduced software-defined radio gear that Ars Technica is comparing to the Apple I. Why? Because software radio can broadcast and receive nearly any radio signal on nearly any frequency at the same time, and thus could 'revolutionize wireless.' The Per Vices Phi is one of the first devices aimed at the mass hobbyist market to take advantage of this technology."

Comments

News Release

[Anonymous Coward]

gnuradio just wasn't sexy enough, I guess. Not enough like arduino on the tip of everyone's tongue.

Re:News Release

[fuzzyfuzzyfungus]

gnuradio just wasn't sexy enough, I guess. Not enough like arduino on the tip of everyone's tongue.

The device in TFA is a piece of hardware designed to support gnuradio(it might support other things as well; but a gnuradio interface is explicitly mentioned in the device specs).

Gnuradio is just the software side. Traditionally, the USRP has been the peripheral of choice. Not cheap; but configurable for a wide range of frequencies and probably the most mature. A sound card(with appropriate external circuitry bringing things down to audio frequencies, of course) is also an option, and certain flavors of DVB TV receiver dongles are the new hotness in the cheap seats.

This "Phi" device lacks some of the versatility of the classier USRP gear; but it is cheaper and offers a very fast interface to the host computer...

Unlike 'Arduino', where the term refers more or less interchangeably to both the software development environment and to a variety of atmega-based microcontroller boards, Gnuradio is just the software side. There is no 'gnuradio' hardware per se, as there is with Arduino. The USRP is probably the closest to being that; but it is pricey enough to be out of the hands of a great many hobbyists.

Re:News Release

[ceoyoyo]

You can get start playing with Arduino for about $25. Less if you can program your own microcontrollers. GNURadio has so far been a software solution with no, or very expensive, hardware to go with it. Not that the Phi changes that, but it's a step in the right direction.

When you can buy a GNURadio setup, including hardware, for under $100, then it'll take off. When it gets down to $25 it will be as popular as Arduino.

Re:News Release

[fuzzyfuzzyfungus]

It's receive only, and the quality isn't magic by any means; but you can get an RTL2832-based DVB-T dongle for ~$20 and be on your merry way.

(And, indeed, this does seem to have spurred greater interest among people who weren't in for a USRP; but were interested. The fact that SDR involves substantially more nontrivial math than many arduino projects probably limits the mass appeal some, though.)

Re:News Release

[Anonymous Coward]

gnuradio just wasn't sexy enough, I guess. Not enough like arduino on the tip of everyone's tongue.

IS this Company trying to say SDR is new and their invention per chance cus if they are they are in BIG trouble radio hams have been using SDR for ages now , A mate of mine puts the surface mounted boards together by hand

Re:News Release

[ceoyoyo]

The TV dongle development makes the whole thing much more interesting. I took a look at GNURadio when it was a new project and lost interest because you couldn't actually do anything with it unless you were already a radio engineer with all the (expensive) equipment to make your own hardware.

Still, you have to find one of those dongles. Somebody really needs to design and manufacture a cheap board specifically for radio, like the Arduino people have done.

Re:News Release

[vandamme]

I doubt it. There isn't that much interesting stuff to receive, that isn't covered by hardwired Rx's. TV, radio, wifi. How many people want to listen to ham radio transmissions, or radar? I've built receivers for 42 years for everything from 60 KHz to X band, for (mostly radar) projects that needed to use those frequencies. For the home hobbyist, it would be interesting for about 3 seconds. Arduinos can do tons of useful stuff.

OTOH I might find a use for them at work.

Re:News Release

[ceoyoyo]

Software defined radios can transmit as well....

Pirate radio?

[Anonymous Coward]

Is this legal?

Re:Pirate radio?

[Anonymous Coward]

Uh, wouldn't that depend on what bands you transmit on, what license you may or may not hold, and what the regulations have to say about it?

Re:Pirate radio?

[Anonymous Coward]

Shhh...don't inform the RIAA/MPAA

Re:Pirate radio?

[ewanm89]

This is only legal to Amateur radio operators I believe, as it will not be vetted by the FCC/Ofcom... to broadcast only on the allowed part of the spectrum. From a Ham operators point of view, it's not much difference than a full HF/VHF rig, or a USRP with every possible daughter board installed. The big change is that it's all on one card and it uses PCI-Express for the interface to to the computer.

The whole this will stop the FCC having so much control over the spectrum is totally wrong. They'll have as much control as they currently do which is they can shut down any transmitter in the US if it's not following the licensing terms they set out.

Re:Pirate radio?

[Iceykitsune]

This is a RECEIVER ONLY! The FCC has said that anyone can listen on any frequency.

Re:Pirate radio?

[Ellis D. Tripp]

Then why are analog cell phone frequency bands blocked on scanners?

Re:Pirate radio?

[SealBeater]

Hmm, not quite as sexy if you can't transmit.

Re:Pirate radio?

[jiriw]

Technical Specifications and Support:

Dual channel, 16 bit, 250 MSPS DAC

Nope ... RX/TX. If you'd read the article by the way, it should have been clear they wanted a SDR that goes both ways. Of course it doesn't really have an amplifier to speak of, so you can't just hook up an antennae to it and expect to work the world. Especially if you want to output multiple signals in multiple bands, as mentioned in the article, things can get very hairy at the transmission end. By the way, good, distortion free, broadband amplifiers aren't cheap as well and come with their own set of problems.

The idea might be nice, an 'open source' spectrum, and for the receiving end it's all fine and dandy. I'm not a proponent, of security through obfuscation/obscurity, so regulation of waves receiving: Governments, just grow up!

But even at low power conditions, for certain frequencies, you don't want to have transmission capabilities in the wrong hands (read: someone who hasn't at least got a a HAM Radio license. A degree in Electronics, Electromechanics, Physics might suffice as well... if it has covered the correct subjects). Things can turn out very nasty even at low power situations. Things like GPS will stop working, or other satellite signals jammed. Many satellites only transmit at an order of 10-100 watts. The amount of signal left when received on earth is miniscule. A little more power and things like Wifi and RC toys/remote controlls/bluetooth will be affected. Digital broadcasting is next I think... including mobile phones and portophone systems there isn't nearly as much robustness in there as there was with the old analogue signals... As they digitized the signals, they could cut bandwith and power requirements... Nice for energy savings and miniaturization of systems but it does mean it can be jammed easier, even if there is overhead in the protocol for error correction. Well .. you'll get the picture.

While I agree this sounds like a better deal for radio enthousiasts than the Ettus USRP and I'll be itching to get my hands on one of these, color me sceptic about the whole low power broadband broadcast 4 noobs vibe the Per Vices founders seem to transmit.

73. PG8W.

Re:Pirate radio?

[zippthorne]

Cell company lobbyists and congress drones apparently thought that you can keep the radio communication equivalent of shouting across a quiet room private by, instead of encrypting the communication, passing laws that make it illegal to notice.

I'm not sure that the FCC or ITU had any part in it, however it seems likely that at least ITU would have been involved....

The remaining question is whether we'll see the law rolled back now that it's been obviated by encryption (or at least CDMA spread spectrum), and is so obviously useless - the only way to detect if someone is listing is to yourself be listening in enough places to measure the shadow created by their receiving equipment or the extremely low-power emission of interference frequencies, assuming that they're using that method of demodulation, equipment capable of receiving the old AM cell phone transmissions can be made in an afternoon using readily available components.

Re:Pirate radio?

[Anonymous Coward]

Not supposed to use Amatuer Radio for music technically

Re:Pirate radio?

[Anonymous Coward]

By the way, good, distortion free, broadband amplifiers aren't cheap as well and come with their own set of problems.

Then what about using the input as a feed-back loop ?
I mean, if you have a device emitting in "all" frequencies (all which matter to your communication), I would expect you can use it to correct distortions by adapting.

Re:Pirate radio?

[Anonymous Coward]

But significantly more legal unless you're a HAM (depending on the band being used)

Which I am and we've been doing this for some time already. This just makes things easier.

I would really advise you and anyone else interested in transmitting or receiving to look up a local HAM radio club and start reading. It's a great hobby for tinkerers and experimenters. Socially the community tends to be significantly less elitist than many other "hackerspace"-y clubs and imnsho much more knowledgeable. Besides, if you are to use something like this effectively, the experience you'll need for the radio amateur license will serve you well anyway so it's a win-win.

Re:Pirate radio?

[vandamme]

Same reason they grope grandmas at the airport. Security theater, legislated by politicians.

Eh?

[OverlordQ]

Most significantly, the widespread adoption of software-defined radio hardware could undermine the FCC's control over the electromagnetic spectrum itself.

No, no it wont. The FCC will bring down the banhammer. If you cause issues, they *will* be knocking on your door.

Right now, the FCC largely focuses on limiting the transmission frequencies of radio hardware. But this regulatory approach is likely to work poorly for software-defined radio devices that aren't confined to any specific frequency.

Yes, yes it will. You cause issues, FCC gets complaints, it sends in the goon squad to shut you down.

Re:Eh?

[icebike]

Not really.
There are large blocks of spectrum already set aside for use of personal radio devices. Just about anything goes in those bandwidths, subject only to power limitations and staying inside of the spectrum block.

The FCC is all for this type of use. The FCC is also fully in favor of reallocation spectrum when the situation and demand changes, which is why analog TV is a thing of the past.

There is precedent for this.

Re:Eh?

[Sponge Bath]

...they *will* be knocking on your door.

Black van pulls up and discharges a swat team:
[*bing bong*]
Resident: Who's there?
Guy in black body armor: "Pizza man!"

I looked at the Ettus Research hardware for a while with the thought of experimenting, but my life is already saturated with work and tech. Software radio will remain alongside playing the guitar as something cool I wanted to do, but could not squeeze in between /. postings.

Re:Eh?

[BlueStrat]

Yes, yes it will. You cause issues, FCC gets complaints, it sends in the goon squad to shut you down.

No, the FCC field operations are a joke. They have been for many years. Budget cuts have all but neutered what little FCC field-monitoring & enforcement that did exist. Many of the monitoring facilities have been shut down or turned into unmanned remote-operated stations.

They've typically got two or three men and one or two tracking vans for a multi-State-wide area. They're kept so busy tracking things like interference to first-responder/aircraft/military/commercial broadcast that most stuff gets a report filed and that's about it.

Strat

Re:Eh?

[osu-neko]

Software radio will remain alongside playing the guitar as something cool I wanted to do, but could not squeeze in between /. postings.

Your priorities... need prioritizing. :p

Re:Eh?

[grumling]

Recent story to back up what OverlordQ said:
http://www.arrl.org/news/california-ham-cited-for-operating-unlicensed-radio-station

Re:Eh?

[Anonymous Coward]

Actually, you are only permitted to operate certified Type 15 devices on ISM bands as per FCC 47 CFR Part 15.5, likewise for PRS UHF transmission, though I can't find the exact regulation that applies to PRS.

So while you could generate type-compliant emissions with the Per Vices Phi, such use would not be considered legal under FCC regulation without additional licensing. Applying for a radio test and measurement laboratory license and installing appropriate RF shielding in your lab would be a sensible course of action. It might also be possible to acquire a field test license.

Re:Eh?

[Anonymous Coward]

Of course, if you fuck with a cellular network or a broadcast operators ST links, they WILL jump in their van/helicopter and track down your transmitter, and when they do, they will file all the documentation with the FCC, then all the FCC agent has to do is serve the warrant.

At least that is how things work in my third-world country (New Zealand). The big networks police their own spectrum, and the MED receive notice of interference and intervene. It would greatly surprise me if the FCC ignored an interference report with complete field testing data.

Re:Eh?

[Anonymous Coward]

So how is a receiver supposed to cause problems? (Hint to only you: a receiver sucks stuff in *ONLY*, a transmitter sends stuff out, they can only tell you are even there if you have a transmitter...). They can't find receivers.

Re:Eh?

[seringen]

That's RIGHT next door to Travis Air Force Base. It is hardly surprising that they noticed.

Re:Eh?

[Anonymous Coward]

Guy in black body armor: "Pizza man!"

They'll deliver a pizza if you tell them to, from their own radios. Mwahaha.

Re:Eh?

[BlueStrat]

Of course, if you fuck with a cellular network or a broadcast operators ST links, they WILL jump in their van/helicopter and track down your transmitter, and when they do, they will file all the documentation with the FCC, then all the FCC agent has to do is serve the warrant.

At least that is how things work in my third-world country (New Zealand). The big networks police their own spectrum, and the MED receive notice of interference and intervene. It would greatly surprise me if the FCC ignored an interference report with complete field testing data.

I was an active US Amateur Radio Service operator ("Ham") for many years as well as being a professional RF/radio communications/avionics/military systems and general electronics technician for about four decades, so I have some idea of which I speak.

There's an FCC monitoring station about 20 minutes away that has three guys and two vans, because one van is usually being fixed. At least, that's how it has been for many, many years, unless DHS, FCC, or somebody gave them a much bigger budget in recent years and I hadn't heard. They cover this state and portions of three others. This state alone is only about ~6,000 sq. miles less area than all of New Zealand. For "three men and a truck". You have to be quite a "sore thumb" to attract their attentions.

Ask US CB'ers (yes, there are still some left) how much of a real threat the FCC is. I'd wager more than half of CB'ers are operating in violation with things like transmitter linear power amplifiers ("footwarmers" or "kickers") boosting the CB transmitter from 4 watts to hundreds, even thousands of watts, "out-of-band" frequency transmission capability, etc. Take-downs are extremely rare, and were even at the height of the "CB craze".

Unless multiple serious complaints of interference with public and commercial services are received, not much if anything will be done outside of possibly receiving a warning letter. If continued complaints are received and you've failed to respond to the warnings, then, your case may get put into the hopper for a mobile monitoring visit. Which, naturally, may easily be three or more months, depending. Unless the violation is of a nature that doesn't require a mobile unit dispatched, where you'll likely receive a summons and/or a warrant delivered by a Federal Marshall, with possibility of arrest and equipment seizure.

Of course, if you're interfering with aircraft/police/fire/military/cell services creating a public safety hazard and endangering lives, resources will be "reallocated" to find and shut you down ASAP. Interfering with AM/FM broadcast stations and OTA TV (like running a pirate station) will also get you on an enhanced-priority list.

They've got a LOT on their plate for the manpower they've got, to put it mildly, so as long as you're not being a complete ass, not causing good people problems and endangering lives, and not attracting attention to yourself, they've got bigger fish to fry.

Now, let me qualify my statements about US FCC attitudes, in that my comments were based upon ~15+ yr-old conversations I've had with field personnel, not FCC Administration officials, and do not reflect (I'm sure) the Administration's attitudes, policies, or official practices and procedures.

But of course the one saving grace is that most of those types of field personnel at the monitoring stations and in the vans are geeks and nerds as well, and probably ham operators too. Most of these guys (not all, sadly) "get it". As usual, it's the suits that are the problem, not the guy wearing a pocket-protector with a field-strength meter in his hand.

Strat

Re:Eh?

[Anonymous Coward]

As one of the pocket-protector wearing R&S equip toting "nerds" who also happens to be a licensed radio amateur myself, let me ask where this inherent hostility towards the FCC comes from? Where I work (not the US) our activity is largely welcomed and serves the purpose of making sure harmful interference is kept to a minimum which is something everyone is interested in.

Is it just the tired old religious Govt-can't-do-anything-right mantra or has the FCC really been that much of a tyrant?

Re:Eh?

[BlueStrat]

As one of the pocket-protector wearing R&S equip toting "nerds" who also happens to be a licensed radio amateur myself, let me ask where this inherent hostility towards the FCC comes from? Where I work (not the US) our activity is largely welcomed and serves the purpose of making sure harmful interference is kept to a minimum which is something everyone is interested in.

Is it just the tired old religious Govt-can't-do-anything-right mantra or has the FCC really been that much of a tyrant?

No hostility here. At least, not towards field personnel that are, as you say, just out there trying to make sure things work. I thought I was quite clear. If anything, I think the field-operations side is woefully under-funded and under-staffed. Nice furniture and a fat budget in the FCC head's offices though, I'm sure.

Any hostility I have is toward the political side of the Commission and for Congress for involving the FCC in such nonsense as "wardrobe malfunctions" and "Fairness Doctrines" and the like.

Although I believe that government has grown too large, there are certain functions that need to be served, and keeping order in the airwaves is one of them. Sadly however these days, far too often essential parts and functions of government go wanting while parts and functions of government not authorized under the Constitution find themselves awash in funds and staff in order to further political and "crony-capitalism" goals and agendas.

Strat

Re:Eh?

[Nethead]

The other thing they do is respond to complaints about self-oscillating BDAs from the cell phone techs that actually find them. I got a service call to meet the FCC at a gas station because the WiFi amp in a pump top TV was going whacky.

73 de w7com

Re:Eh?

[Anonymous Coward]

I concur, the field work is important and underfunded. Meanwhile the political crowd at the top also has a tendency to push agendas that make no technical or business sense and threaten existing licensed uses. Two examples are the insane push the FCC made for broadband over powerlines that turned overhead power lines into massive antennas spreading interference hither and yon while providing essentially no public benefit. Another example was the recent boneheaded attempt to fast track the use of satellite frequencies for broadband that threatened GPS use. These both should have had zero traction, yet were actively pushed by the administrations at the time.

Re:Eh?

[Muad'Dave]

Not true. All direct conversion and superheterodyne receivers contain circuitry that oscillates at the receiver's IF frequency. A very small amount of that RF leaks out and can be detected from nearby. Radar detector-detectors use this technique.

The story goes that the TV van in England uses IF detection to find TVs that are not licensed, but much doubt has been cast on this story.

What we're on the Radio now?

[Anonymous Coward]

This... is my voice... on the R-A-D-I-O!

URL to Printable Article?

[Anonymous Coward]

Does anybody have an URL to a printable copy of the article? I couldn't find one.

USRP is expensive

[chihowa]

The USRP is really cool, but stupidly expensive. Some really cool stuff is happening with the RTL2832 based TV dongles, though. These are $20 devices that can be used to receive from ~64-1700 MHz (or DC-30ish with a little tweaking). So far, much of the info is here

The USRP would be cool if current PCB layouts and schematics were available or if the development effort went to a system that wasn't just making Ettus a profit. A truly open development platform would really benefit the SDR community.

Re:USRP is expensive

[SealBeater]

I have a USRP2 and I agree, the cost is prohibative, but to be fair, some of the SMD parts are as much as $45 bux each, and it's a 8 layer PCB.

Re:USRP is expensive

[Anonymous Coward]

Prohibitive?

LIke tbe B100+WBX bundle for $849.00? That's only a little more than the Per Vices device, which I'll point out, has no market history, and they have what I
    would have to describe as "kindergarten" Gnu Radio support.

Compare that to the latest gee-whiz ICOM/YAESU/KENWOOD does-everything-with DSP HF-only "rig", and it's dirt cheap. I get frustrated by ham-radio guys
    (and I'm one) balking at "computer stuff is too expensive", and then a month later at the club meeting showing off their latest only-does-one-thing $3000.00
    "appliance".

In the market where the Ettus Radios compete, they're considered *cheap, cheap, cheap*. That market is the academic and industrial wireless development
    market. Perhaps a bit pricey for the budget-constrained hobbyist, but cheaper than those "appliances" so many hams are very fond of trotting out in a
    penis-size competition with their similarly-inclined brethen. And much much cheaper than the "incumbents" in this industry -- Rhode & Schwarz, PenTek,
    etc, etc.

Re:USRP is expensive

[Anonymous Coward]

Yeah, because, companies making profits from hardware are evil beyond description.
Why, perhaps we should all move into tin shacks and have all the non-profit consortia provide hardware at cost to everyone. That would be just the Utopia I
    imagine folks like you are looking for, correct?

Or people should make all hardware and software at no profit to themselves, and hunt wild animals to feed their families, and fell trees to make their homes, and
    contrive buckets from bearskins to haul water. But hey, at least all their software and high-tech stuff will have been purchased at cost. With what, I'm not sure.
    perhaps traded for bearskins.

Re:USRP is expensive

[thygate]

The RTLSDR is nice, but has VERY poor dynamic range. (8 bit I/Q samples), and no Tx capability. I am however having lots of fun with one, since i can't fork the $1400 for an USRP.

Re:USRP is expensive

[Anonymous Coward]

Instead of giving him shit maybe you should recognize that there is a market not being served by Ettus' profit maximization and that the TV Tuner dongle frenzy is an example of smaller margins making equal sums of money for the inventor, but to the benefit of a larger number of people.

Instead of being sarcastic you should encourage him to take out a bunch of student loans, go back to school to study EE, and then upon graduation start a kickstarter project where he sells a USRP tier device for $200.

Re:USRP is expensive

[Anonymous Coward]

You realize that the RTL dongle wasn't even *designed* as an SDR device, right? It was designed to service mass-market, consumer DVB-T and DAB reception.
    The economies of scale in the consumer electronics market just don't compare at ALL to what still amounts to a niche market--SDR "platforms" for
    commercial research and development, and academic research, and the occasional hobbyist.

The RTL-SDR "phenom" is entirely accidental. Like discovering that the your $10,000 Ford car actually has an engineering mode that purely by happenstance
    allows you to travel at the speed of light--and accidental billion-dollar feature.

There's a *huge* difference between a SDR platform that *was engineered* to be a general-purpose SDR platform, and one that sorta-kinda works as a "toy"
    some of the time. A bit like complaining that you should have to pay $10,000 for your Ford vehicle, when you bought a used bicycle just the other day
    that gets you from A to B for only $20.00. Surely Ford could learn a thing or two from the bicycle manufacturers?

Re:USRP is expensive

[Anonymous Coward]

A bit like complaining that you should have to pay $10,000 for your Ford vehicle, when you bought a used bicycle just the other day that gets you from A to B for only $20.00. Surely Ford could learn a thing or two from the bicycle manufacturers?

I have a feeling you don't live in a city do you?

Very little to do with broadcast

[icebike]

The game changer here would be in the Cell Phone industry which can substitute a single radio chip to do all the protocols, wifi, cellular, bluetooth, as well as mix and match them at will. New air protocols could be invented over night without waiting for expensive chip developments. Its a cost reduction path as well as a device longevity path.

Although it sounds wonderful when your cell phone is stuck on CDMA or your Bluetooth lacks all the latest capabilities, there are still problems of having an infinite number of antennas available (yes, we already have software defined antennas) in a small place.

There will still have to be frequency restrictions imposed in the hardware itself because the FCC can't afford to allow Joe Random Programmer bringing down jumbo jets. But within authorized bands the ability to use new methods without waiting for the next chip means that we can build a replacement for entire infrastructures much more quickly, while maintaining existing technology for as long as we need it.

Somewhere in this world there are still 029 card punches in use. I suspect we will keep some of our current stuff long after it should be scrapped.

Re:Very little to do with broadcast

[ceoyoyo]

I don't think it's going to do much for cell phones. Software defined radio basically shifts the processing from hardware to software. That requires power. For a cell phone, which must operate on a set protocol, there are only drawbacks. Yes, you could upgrade the protocol, but cell protocols don't change very fast and it's unlikely you'd want to run a general purpose cell tower on SDR because of the processing requirements.

What SDR is going to do is revolutionize the unlicensed bands.

Re:Very little to do with broadcast

[ewanm89]

No, they won't, they lock the chips to specific protocols for a reason, wifi and bluetooth chips are already on the same band and could be combined with very little work, but all the work the vendors do to lock the chips to one specific protocol it's insane. Unfortunately the FCC/Ofcom/ITU regs pretty much say they have to, this is why all wifi cards have some binary blob somewhere (firmware upload to device, firmware already on device...) to stop you accessing all the frequencies the hardware is capable of or with non-certified protocols and yet allow the same hardware to be used in multiple countries with differing regulations.

Re:Very little to do with broadcast

[sir-gold]

I'm sure there are boatloads of wifi/BT/2g/3g/4g patents standing in the way of this.

The recent ITU lawsuits have shown that it's nearly impossible to build even a single-purpose wifi chip without stepping on SOMEONE'S patents, let alone a multi-protocol chip

Re:Very little to do with broadcast

[Baloroth]

In theory, though, an SDR cell phone could transition from 3G to 4G-LTE to true 4G with nothing but a software update. That is an extremely cool idea. Tt'd also allow fancy things like using it as a true walkie-talkie or CB radio, and 100% world-wide compatibility. I agree it is unlikely to happen anytime soon, but TFA compares the current SDR systems to the Apple I: it's going to take a very long time before the technology sees it's full usage.

Re:Very little to do with broadcast

[ceoyoyo]

Most phones already use chips that have wifi, bluetooth and GPS in one.

Re:Very little to do with broadcast

[icebike]

Actually I think it will happen VERY soon. Within a year or two.

Why? Because there are so many different radio standards in Cellular use already, in so many different Frequency Blocks, and so many different protocols.
Handset manufacturers would love to have one radio package to install and be done.
Who ever comes out with one of these that can be switched to handle any cellular network worldwide with just an API call wins. Its game over for all the discrete chip makers.

Plus if you can strip out the WIFI, and Bluetooth chipset there will be more power savings. And the software to do this won't be any more trouble or require any more power than the discrete chips do now. (Radio chipsets are already run by software. We call them binary blobs in linux, or Roms in Android.).

Re:Very little to do with broadcast

[ceoyoyo]

In reality, though, a phone that was able to do that would have to have a processor big enough to handle the extra overhead from decoding 4G, would burn through it's batteries in no time, and would be more expensive. And you wouldn't be able to do CB because you wouldn't have an appropriate antenna, although you probably could turn it into an FRS walkie talkie. There are already multi-frequency CDMA/GSM world phones chips. To support CDMA and GSM you need to have the identity module hardware for both anyway.

It might happen someday, but I think things like cell phones are the LAST place SDR is going to be used. Cell phones are already one of the only places you'll find dedicated media decoding chips, and for the same reason.

Re:Very little to do with broadcast

[MobileTatsu-NJG]

In theory, though, an SDR cell phone could transition from 3G to 4G-LTE to true 4G with nothing but a software update.

Please forgive my ignorance on this topic, but wouldn't processing power on-board the device still be a limiting factor? Is it possible that to leap from 3g to 4g that you'd have to get something with a much faster processor? Or is this the sort of thing where the processors are already fast and cheap enough?

Re:Very little to do with broadcast

[AmiMoJo]

Phones already use SDR. It is built into the radio ICs which contain both the analogue radio hardware and a programmable DSP. Android phone updates often include firmware updates for these DSPs, and many phones can be swapped between W-CDMA/GSM and whatever it is the US uses just by changing the radio software.

Re:Very little to do with broadcast

[Bob The Cowboy]

Software defined radio basically shifts the processing from hardware to software.

So what you're saying is that we need hardware accelerated software defined radios?

Re:Very little to do with broadcast

[jiriw]

However ... if you could just use a generic hardware broadcasting device and do all the patent-laden de/encoding in software... You'd have a blast in those large regions of the world where software patents don't hold much sway (Europe, for example... 'though lobbyists try to change that quite vigorously).
And when the U.S. finally learns 'idea' patents only hamper innovation, there won't be a problem at all. It'll be 'just' software ;)

Re:Very little to do with broadcast

[Anonymous Coward]

Cell phones do already use SDR, the only distinction seems to be that in a cellphone the RF frontends are tuned to 1-5 fixed bands as it allows better performance at lower cost and power. They radio simply switches filter or filter-configuration to switch between 850, 900, 1800, 1900 and 2100 Mhz, but then it feeds into the same 50MHz IF stage and ADC, where software picks out a channel and demodulates it.

Re:Very little to do with broadcast

[shiftless]

In reality, though, a phone that was able to do that would have to have a processor big enough to handle the extra overhead from decoding 4G, would burn through it's batteries in no time, and would be more expensive.

Not if you used a special purpose (programmable) radio processor (instead of the CPU) to do the software decoding.

Re:Very little to do with broadcast

[Anonymous Coward]

In reality, though, a phone that was able to do that would have to have a processor big enough to handle the extra overhead from decoding 4G, would burn through it's batteries in no time, and would be more expensive.

Not if you used a special purpose (programmable) radio processor (instead of the CPU) to do the software decoding.

Like they already do. Some cheap phones are Cheap because they run the shit on a single arm. has to have chips for output amplifying and filters though but you would need those anyways.

Re:Very little to do with broadcast

[ceoyoyo]

One of the big benefits of SDR is that it's able to use any set of frequencies you like. That means you need to have a high speed ADC and the ability to process those samples. That means a decent processor, special purpose or not (they use FPGAs in the boards mentioned in the article). Compare that with the simple oscillator and mixing circuits that are usually used.

Sure, you can use a DSP or other special purpose chip to produce the baseband signal (and that's what's often done), but that's not really what SDR is about.

GNU Radio needs hardware behind it

[tepples]

The article mentions GNU Radio, saying that the hardware used with GNU Radio during the "broadcast flag" debate couldn't capture more than 0.1 MHz of spectrum.

Re:GNU Radio needs hardware behind it

[Man On Pink Corner]

The USRP1 used a USB 2.0 high-speed interface (Cypress FX2), and it could certainly downconvert at a wider bandwidth than 100 kHz. It would have had to, in order to demodulate ATSC.

I don't know where that line in the article came from. Reporters making up random stuff as usual, I suppose.

What about signal amplification?

[Anonymous Coward]

What about signal amplification? RF PAs are usually only suited for a limited range of the RF spectrum. How does software defined radio get around this?

so? old hat.

[swschrad]

hams have had SDR for a decade more or less. and software-controlled radio back a little longer. and I seem to remember a win95 radio card that slid into an AT slot back in the mid or late 90s...

Re:gnu radio

[Anonymous Coward]

I don't know why the sockpuppets are badmouthing the GNU so loudly lately. That ship has sailed. They won. Free(as in freedom) software is legitimate, is a legitimate business model, and is eating the lunch of commercial software in many arenas while chipping away at others. Even the BSDers with their doublespeak about the GPL being "less free" amount to little more than a gigantic pile of butthurt. Maybe you could have made that argument a decade ago, but history has also vindicated the FSF here too.

The FSF and Stallman are like.. Socrates. Principled and unwavering, uncomfortably correct. Their detractors have little recourse but to badmouth them. (Haha communisim, haha dirty bearded hippy.)

Software Defined Radios going to ISS in 15 days

[ajalics]

Interesting time to talk about Software Defined Radios.

NASA's SCaN Testbed with 3 Software Defined Radios is launching onboard the Japanese HTV-3 Unmanned cargo vehicle in 15 days. (July 21st)

It's an experimental payload that will be bolted to the exterior of the International Space Station and perform communications experiments with the 3 SDR's contained in the payload.

http://spaceflightsystems.grc.nasa.gov/SOPO/SCO/SCaNTestbed/Payload/

Re:Software Defined Radios going to ISS in 15 days

[Anonymous Coward]

July 21st in Japan..
July 20th in the US

And what's more cool is that Joe Schmo can propose to try something, and if it gets through the review process, you too could have software flying on a radio on ISS, at no cost to the experimenter (beyond their own time). Not that lots of individuals are expected, but certainly, university projects and such.

Only half of the widget...

[Worchaa]

FTFA: "It could record FM radio and digital television signals, read RFID chips, track ship locations, or do radio astronomy. In principle it could perform all of these functions simultaneously."

Nice try, but no. At least not in a practical sense and certainly not as a mobile rig.

Software Defined Radios are sweet but still dependent on a Physically Defined Antenna. I can see loads of wonderful uses for a broadband, frequency-agile SDR. Actually, I use them often as a Ham radio operator and they are extremely cool. However, there's still the problem of the pesky antenna. You can fudge quite a bit on a receiving antenna, not so much with a transmitting antenna (or a single transceiver antenna), and the engineers out there are very talented and clever at coming up with better designs... but it always tends to come down to the antenna.

My point is that advances in SDR tech is fantastic, but they're not-- nor do I ever see them becoming-- a magic box. What I think they WILL do is streamline production. One super SDR can be dropped into a number of application-specific boxes.

Re:Only half of the widget...

[osu-neko]

You can fudge quite a bit on a receiving antenna, not so much with a transmitting antenna...

None of the uses in the quote you objected to require a transmitting antenna.

Re:Only half of the widget...

[Worchaa]

You can fudge quite a bit on a receiving antenna, not so much with a transmitting antenna...

None of the uses in the quote you objected to require a transmitting antenna.

True. That's a good point. However, consider the enormous range of those services:

RFID: 120 KHz - 10 GHz (Generally below 2.4 GHz, with LF and UHF tags being common)
FM Broadcast: 88 MHz - 108 MHz
DTV: 55 MHz - 700 MHz (Three bands, ~55-85, ~175-210, ~470-700)
Radio Astronomy: 13 MHz - 0.8 THz or something equally nuts way up there (The VLA receives below 50 GHz)

That's way outside the scope of getting an antenna to fudge on receive. We're talking wavelengths from ~1.5 MILES to under half a millimeter !

Unless Scotty beams down and hands us an antenna from the future, TFA's mega-broadband SDR described as doing all that at the same time is science fiction. I'll buy that SDRs could possibly handle that kind of bandwidth sometime soon, but there's no way we're going to see a practical antenna system shipping with it.

Re:Only half of the widget...

[Anonymous Coward]

Uh, software defined antennae exist. steerable phased array antenna are a common example, but broadband tunable admittance band antennae also exist.

Re:Only half of the widget...

[Muad'Dave]

There are better ways to accomplish what you want - you can switch quickly between dedicated receive antennas to obtain samples across all those bands effectively* simultaneously. Frequency agile receivers do that all the time.

*We're talking discrete sampling here, so as long as you can switch antennas (PIN diodes) as fast or faster than the (possibly under-) sample rate and the settling time of the receiver when the frequency is changed, the receiver will have no idea there's more than 1 antenna. As the English say, "Bob's your uncle!"

JTRS

[zjbs14]

The DoD should have kicked in a few bucks to this project instead of wasting billions on, and then cancelling, JTRS (Joint Tactical Radio System).

Re:JTRS

[Archangel Michael]

I don't think you understand the purpose of programs like the JTRS. The weren't supposed to actually make anything that was functional. The program was designed to make the contractor a shit load of money off the US Government.

Re:JTRS

[Anonymous Coward]

I don't think neither of you understand the cost behind porting legacy/propriatary/new waveforms onto CORBA/SCA compliant SDR platforms.

Creating a SDR platform is easy, if you want the performance of and inter-operability with legacy waveforms and equipment that have ASICs and may rely on platform specific quirks outside waveform standards... well then, you better pony up the cash.

Re:JTRS

[Man On Pink Corner]

The whole idea behind SDR is that anything that formerly required an ASIC can now be done on general-purpose CPU, and/or an FPGA.

You can pretty much rest assured that there is nothing in any legacy system used by DoD that couldn't be implemented on this device, or on a USRP2.

Re:JTRS

[Anonymous Coward]

That's the idea... doesn't mean it'll actually work out in practice.

DoD want's a SCA compliant, Type 1 cryptographic architecture for it's most secure systems. There goes USRP2 out the window right away. But let's ignore that for now.

JPEO JTRS program has ~4million lines of code.

On top of that, these radios are connected into legacy systems that often have stringent timing requirements tied to the legacy radios in order to operate properly.

This comes close: http://rf.harris.com/capabilities/tactical-radios-networking/an-prc-117g/

But it only fills out a minor percentage of what JTRS is supposed to accomplish. On top of that, my bet is that their SINCGARS waveform isn't 100% compatible with this: http://www.exelisinc.com/solutions/SINCGARS-RT-1523/Pages/default.aspx

It may seem to, but remember there's are lots of proprietary and classified specs that you'll never see.

Soooo... my bet is no, USRP2 wouldn't come close to filling everything the DoD wanted.

Re:JTRS

[Man On Pink Corner]

The four million lines of code wouldn't run on the USRP2 as such; they would run on the host PC or other system. (When you hear figures like that batted about in this context, it's usually because they're referring to a collection of every codec and encryption standard used since the Spanish-American War. The amount of that code that actually needs to run at any given time is much smaller.)

What a COTS SDR would bring to the table is hardware reuse. Things like cryptographic architectures belong at other layers of the OSI hierarchy; they should have nothing to do with the hardware spec.

Maybe I'm wrong. Maybe I'm all wet. Or maybe I'm not the guy who spent $20 billion on radios that either don't exist or don't work.

Re:JTRS

[Anonymous Coward]

COTS SDR wouldn't bring anymore hardware re-use than what is already fielded.

Let's say you use a USRP2 for the front end. You don't get much (or even any) more capability than the AN/PRC 148, AN/PRC 117G, AN/PRC 152. You're also missing the cryptographic component, battery life, rugedness and any ability to configure the radio on the fly.

Also, I'm not sure why you're bringing up OSI here, these are custom military systems running waveforms that have been developed as early as the late 70's, not standardized commercial systems. A Type 1 cryptographic / SCA compliant radio is... well a Type 1 cryptographic / SCA compliant radio. I suppose you can use a COTS SDR part on the front end, buy a crpytographic chip from ITT or GD and stick it in a seperate box cabled to your COTS SDR part and then have another box cabled to that box to handle the HMI and unencrypted traffic processing. Now you're almost to the point where you can compete with the radios I mentioned above which are already fielded.

You are however still missing the software portion of the radio, whatever % of the 4million lines of code you chose to port. After that, it's just a *short* stretch through JTIC/IOP testing and NSA approvals and you're done...

As I stated before, creating a SDR platform is easy, getting all the waveforms to run on the on the platform and still inter-operate properly with legacy radios and systems is going to cost you and arm and a leg.

As for the radios that don't exist. See the three I mentioned above. A good portion of the $6billion pricetag can be attributed to gross under-estimation of the complexity of the task and feature creep. The good part of it, is it generated a lot of research and knowledge that private contractors can tap into and create high end, function SDRs.

Methinks you're just overly optimistic about COTS SDR parts for the military and underestimate the engineers that actually make the stuff ;-).

The FCC heavily regulates SDRs

[TwineLogic]

It would be good to change the laws and federal regulations in the United States so that using SDRs would become legal. The current situation is an attempt to enforce "privacy through obscurity" by outlawing radios which could possibly intercept cell phone, pager, or radiotelephone communications (47 USC 302). It is also an attempt to enforce "copyright through obscurity" by requiring that FCC-approved devices respect copyright bits (47 USC 605). All of these problems would be better solved with cryptography. Remember the Clipper chip? That would have been a better path to choose than the current situation.

A few of the relevant obstructions in the form FCC regulations and laws are: 47 USC 2.501, 47 USC 302, 47 USC 605, 47 CFR 2.944, 47 CFR 15.3 (dd).

47 CFR 2.944:
Software defined radios.
(a) Manufacturers must take steps to ensure that only software that has been approved with a software defined radio can be loaded into the radio. The software must not allow the user to operate the transmitter with operating frequencies, output power, modulation types or other radio frequency parameters outside those that were approved. Manufacturers may use means including, but not limited to the use of a private network that allows only authenticated users to download software, electronic signatures in software or coding in hardware that is decoded by software to verify that new software can be legally loaded into a device to meet these requirements and must describe the methods in their application for equipment authorization.
(b) Any radio in which the software is designed or expected to be modified by a party other than the manufacturer and would affect the operating parameters of frequency range, modulation type or maximum output power (either radiated or conducted), or the circumstances under which the transmitter operates in accordance with Commission rules, must comply with the requirements in paragraph (a) of this section and must be certified as a software defined radio.
(c) Applications for certification of software defined radios must include a high level operational description or flow diagram of the software that controls the radio frequency operating parameters.
[70 FR 23039, May 4, 2005]

The penalty for a violation is forfeiture, a fine of up to $10,000, and up to one year in federal prison. See 47 USC sec. 501 This applies to person who purchase the radios as well as persons who sell them. See 47 USC sec. 500 et. seq.
Various internet sources assert that SDRs are "test equipment" and excluded under 47 CFR 15.3 (dd), which reads:

(dd) Test equipment is defined as equipment that is intended primarily for purposes of performing measurements or scientific investigations. Such equipment includes, but is not limited to, field strength meters, spectrum analyzers, and modulation monitors.

I find it difficult to believe the FCC would classify the various SDRs as test equipment, but we will probably find out soon enough.

http://www.law.cornell.edu/cfr/text/47/2.944
http://www.law.cornell.edu/uscode/text/47/501

Before you downvote me because you don't like the laws; consider this: I posted this information because we must change these laws rather than suffer them.

Re:The FCC heavily regulates SDRs

[tlhIngan]

There is basically no regulation on SDRs.

Receivers - well, you have normal receive rules, though the cellphone one is pretty much invalid these days as no one uses AMPS anymore.

Transmitters - the rule basically says if you have a software transmitter, that software better only allow transmission on the licensed bands.

There aren't any special rules other than "don't transmit where you're not licensed to". The rule for software options is basically ensuring that the user cannot misprogram their transmitter and operate out of band and interfere with other licensed services.

It's the same as an old style transmitter - care should be taken so users cannot readily change the operating frequency and power so they create interference.

And yes, the fines are like that because they apply to unlicensed transmitters as well - if you're transmitting on a band you're not supposed to, you, the user can find your equipment confiscated and fined.

The law is perfectly adequate - manufacturers need to ensure their SDR cannot be used out of the licensed bands (and power envelopes). It's an "SDR" rule because in an old style transmitter, the output stages normally dictate that you can't transmit out of band anyhow without retuning. But since an SDR can be free to transmit on any band without limitation, the software must ensure it's within the license and the user can't trivially modify it to be out of spec.

SDRs are everywhere - the modern cellphone, wifi radio, bluetooth, etc., they're all SDRs internally. These normally have very specific front ends and filters so even if you could set them out of band, they are out of tune and don't transmit squat.

It's a fair rule and prevents frequency anarchy (and frequencies are set aside for various uses).

Re:The FCC heavily regulates SDRs

[timeOday]

The laws you quoted appear to restrict transmitting, not receiving. As written it seems to me you could distribute "approved" software that would allow anybody to receive anything.

So, has anybody been prosecuted for receiving signals, or distributing equipment to receive a signal? (Short of circumventing encryption?)

Re:The FCC heavily regulates SDRs

[scharkalvin]

There has never really been ANY law against owning or building ANY radio receiver that could pick up ANY part of the spectrum. Scanners have been sold that blocked out Cell phone frequencies, but people have hacked these to re-enable the reception. Today, the point is moot as Cellphones have gone digital and the scanners were all analog receivers. It was ALWAYS illegal to make public any conversations you heard on ANY "public service" radio band, this includes CB, mobile phone, cell phone, etc. (The amateur radio bands are an exception to this). Transmitters, OTHO ARE regulated by the FCC as to what can be sold. Hams may build their own equipment, and they don't really have to meet any FCC regulations (however the quality of the signals, power levels, bandwidth, etc MUST meet regulations). Commerical amateur radio equipment IS subject to some minimal regulation (such as the imfamous 10 meter restriction to prevent their use on 27mhz CB). The power output levels of a basic SWDR transceiver are about as low as a typical rf signal generator. So yes, these devices WILL be regulated as if they were just lab equipment. It's not until you connect them to an antenna that they actually become a transmitter (and you could hook up a typical tv/radio repair shop type rf signal generator to an antenna an go on the air. I know of some bootleg broadcasters that HAVE done this with a QRP station). If you connect such a rig to a suitable rf linear power amplifier you could have a nice ham rig capable of any mode on any band.
It would be futile to regulate SDR equipment, since they are only one means of getting an illegal station on the air. Want to build a bootleg transmitter? Just look for ANY copy of the ARRL radio amateur's handbook for ALL the data you'd need. It's really not rocket science and the cat's been out of the bag almost a century!

Re:The FCC heavily regulates SDRs

[TwineLogic]

You haven't read the laws or the regulations I cited in context. In particular, the laws define "interference" from a receiver as being able to listen to AMPS or to decode a digital cell phone or pager signal. I realize that isn't "interference" in any scientific definition, but the law defines it as such, and that's what will count in court.

Re:The FCC heavily regulates SDRs

[TwineLogic]

Again, you have not read the laws I cited a few of. Receivers which can tune to AMPS are illegal. Receivers must not cause interference, and the definition of interference includes the ability to receiver cell phone signals. See 47 CFR 302A (d):

(d) Cellular telecommunications receivers (1) Within 180 days after October 28, 1992, the Commission shall prescribe and make effective regulations denying equipment authorization (under part 15 of title 47, Code of Federal Regulations, or any other part of that title) for any scanning receiver that is capable of— (A) receiving transmissions in the frequencies allocated to the domestic cellular radio telecommunications service, (B) readily being altered by the user to receive transmissions in such frequencies, or (C) being equipped with decoders that convert digital cellular transmissions to analog voice audio. (2) Beginning 1 year after the effective date of the regulations adopted pursuant to paragraph (1), no receiver having the capabilities described in subparagraph (A), (B), or (C) of paragraph (1), as such capabilities are defined in such regulations, shall be manufactured in the United States or imported for use in the United States.

A key definition is "scanning receiver":

(v) Scanning receiver. For the purpose of this part, this is a receiver that automatically switches among two or more frequencies in the range of 30 to 960 MHz and that is capable of stopping at and receiving a radio signal detected on a frequency. Receivers designed solely for the reception of the broadcast signals under part 73 of this chapter, for the reception of NOAA broadcast weather band signals, or for operation as part of a licensed service are not included in this definition.

I submit to you the legal theory that an SDR receiver is a scanning receiver. I could be wrong, but it would depend on the mood of a judge.

Re:The FCC heavily regulates SDRs

[nurb432]

It's a fair rule and prevents frequency anarchy (and frequencies are set aside for various uses).

Not all will agree.

Re:The FCC heavily regulates SDRs

[jrincayc]

Um, I agree that the regulation prohibits selling a receiver that can receive AMPS cellular service, but the laws you cite as I read them don't prohibit building or owning a receiver that can receive AMPS.

Re:The FCC heavily regulates SDRs

[Anonymous Coward]

If you want your cellphone, GPS, wifi, etc. to work reliably, then you need laws like this.

Re:The FCC heavily regulates SDRs

[TwineLogic]

What, do you want me to the entire Title 47 of US Code? Anyway, I'll save you the trouble of finding this:

TITLE III—PROVISIONS RELATING TO RADIO
PART I GENERAL PROVISIONS
SEC. 302. [47 U.S.C. 302] DEVICES WHICH INTERFERE WITH RADIO RECEPTION.
(b) No person shall manufacture, import, sell, offer for sale, or ship devices or home electronic equipment and systems, or use devices, which fail to comply with regulations promulgated pursuant to this section.
[...]
TITLE V—PENAL PROVISIONS – FORFEITURES
SEC. 501. [47 U.S.C. 501] GENERAL PENALTY.
Any person who willfully and knowingly does or causes or suffers to be done any act, matter, or thing, in this Act prohibited or declared to be unlawful, or who willfully and knowingly omits or fails to do any act, matter, or thing in this Act required to be done, or willfully and knowingly causes or suffers such omission or failure, shall upon conviction thereof, be punished for such offense, for which no penalty (other than a forfeiture) is provided in this Act, by a fine of not more than $10,000 or by imprisonment for a term not exceeding one year, or both; except that any person, having been once convicted of an offense punishable under this section, who is subsequently convicted of violating any provision of this Act punishable under this section, shall be punished by a fine of not more than $10,000 or by imprisonment for a term not exceeding two years, or both.

Again, to reiterate, the ability to receive cell phone signals, pager signals, or copyright-bit-set ASTC is elsewhere defined as "interference." I dunno if I already posted a reference to that, but I assure you, it's in there.

Also, be aware that home-built equipment are excluded, provided they are never marketed, and you build no more than five of them. Kits, however, are not excluded from the regulations.

It's a violation of a different law if you use a home built device to actually listen to a "private conversation," whatever that is when broadcast in the clear. I would assume cell phone conversations (AMPS or digital) and pager traffic would automatically be included in that, as they are explicitly mentioned in the USC.

Re:The FCC heavily regulates SDRs

[shiftless]

Maybe

Re:The FCC heavily regulates SDRs

[shiftless]

What, do you want me to

the entire Title 47 of US Code? Anyway, I'll save you the trouble of finding this:

Fuck you and your laws. Go rain on somebody else's parade. We're trying to advance the state of the art here, not be dragged down by some obscure, forgotten regulation amongst tens of thousands of other laws that time forgot. Take your negative nancy-ism elsewhere please.

Re:The FCC heavily regulates SDRs

[TwineLogic]

Wow are you a 12-year old? The reason I'm posting this is the laws should be changed. Educating the population which is or could be affected by these laws is my effort to change these laws.

Personally, I'd like to sell an SDR, but I can't because of the law.

They aren't my laws, they are the laws of the United States. Live within the borders and ignore the laws at your own peril. It would be better if you tried to change them, rather than name-calling and making an abject fool of yourself.

What are you doing to improve the United States? I assume nothing.

Re:The FCC heavily regulates SDRs

[jrincayc]

Thank you for your posts, they have been useful. Changing the receiving laws has been discussed before such as at http://www.arrl.org/forum/topics/view/112 Physically, there have to be rules on transmitting, since bandwidth is limited. (For example, the entire available radio bandwidth between the maximum usable frequency and the lowest usable frequency between say, South America and Europe is less than a single gigabit Ethernet connection.) I am not sure how the rules should operate in with SDR's.

The Apple II ...

[Anonymous Coward]

The Apple II was sold with a separate RF modulator because of potential RF interference. Software defined radio is RF interference, almost by definition. (Seriously. You have to know what you're doing in order to avoid interference, and that knowledge ain't trivial.)

Re:so? old hat.

[Anonymous Coward]

Yeah - FlexRadio comes to mind whenever SDRs are mentioned.

Apple I?

[narcc]

Er, the Apple I didn't really revolutionize anything. (The Apple II was easily the more influential Apple computer, but even then that was mostly due to VisiCalc.)

Why not "the MITS Altair of broadcast", ars? You know, a computer that had a real influence on the personal computing revolution.

If they just wanted something really early, why not "the Kenbak-1 of broadcast" or "the H8 of broadcast"?

Before everyone accuses me of worshiping at the alter of a dead cult-leader like Roberts, here's what I'm thinking: They picked the Apple I to attract clicks from readers who would otherwise have no interest in software defined radio.

Re:Apple I?

[Anonymous Coward]

MITS?! Other than serving as Microsoft's launch vehicle, the Altair represented an evolutionary dead end. Nothing programmed with toggle switches was ever going to influence the PC revolution, besides bringing a few of the earliest microprocessor hackers on board.

Re:Apple I?

[msauve]

Actually, the Apple I was quite competitive. For $245, you could have a KIM-1, with a calculator keypad and 6 7-segment LEDs, and a slow cassette interface for I/O (you could also connect a TTY and there were programmable TTL ports), and a pretty basic monitor (CLI monitor, not video monitor). It had a bit over 1K of memory, all other expansion was off board. And you had to provide regulated power supplies.

The Apple I, for $667, got you 40x24 NTSC output, easy connection to an ASCII keyboard, on-board voltage regulators, and 4K of memory (expandable to 8K on board).

The Altair was much more expensive, well over $1000 for a usable system (cpu, memory and any i/o other than the front panel), as were the later S-100 systems (the IMSAI, Poly-88, SOL, etc.). It was also huge compared to the KIM-1 or Apple I.

But it was 1977, and the trio of Apple ][, Commodore PET 2001, and Radio Shack TRS-80 when things really took off. Those three made it possible for people with no soldering skills to get involved.

Re:Apple I?

[narcc]

?

I was talking about the computer's influence on the personal computing revolution, not how competitive it was in the market.

Re:Apple I?

[msauve]

...and yet you pointed to the Altair, which really had little influence. Yes, it was first, but left no lasting legacy (S-100 was pretty much dead by the time the IBM PC came out). It's not as if microcomputers wouldn't exist if the Altair hadn't appeared. And where is Altair (or IMSAI, or Polymorphic Systems, or North Star, or Morrow, or Cromemco, etc. now? Clearly, the plain fact that Apple is the largest tech company in the world makes its first product one of lasting importance.

The Apple I was much closer to a modern all-in-one system, you didn't need to add much to be functional. Keyboard and display, about the same as today, and a transformer for power. As I already said, it was the Apple ][, PET, and TRS-80 which made microcomputers pretty much plug and play.

And the real advantage of the A2 was the availability of lower cost, higher capacity, easier to use disk drives, along with the ability to support significant on-board RAM expansion. That in turn made Visicalc practical. Even though it was first on the Apple ][, Visicalc simply didn't work as well on a PET or TRS-80.

Re:Apple I?

[narcc]

...and yet you pointed to the Altair, which really had little influence.

What? It's the single most influential computer of the time! It was the 8800 on the cover of Popular Electronics that inspired Bill Gates and the old Traf-O-Data crew to create a BASIC for the machine and later found Micro-soft corporation. It sparked an entire industry of peripherals and countless companies like CroMemCo to say nothing of clone machines like the IMSAI 8800. It even set the first de facto personal computer industry standard: the S-100 Bus!

Even the famous Homebrew Computer Club was founded in response the Altair. That club had a very heavy influence on Woz, who claims that the club renewed his interest in computers. Would Apple even exist without the Altair and the HCC? I very seriously doubt it!

If there's any computer that can be credited for starting the personal computing revolution, it's the MITS Altair 8800.

Now, MITS was seriously influential itself (though all due to the success of the Altair and the industry it created) pioneering not only an entire industry by releasing the first affordable personal computer, but also virtually every aspect of the early industry including magazines, shows, retail sales, user groups, ... the list goes on.

To say that the Altair hardly had any influence is just absurd. It would be difficult to overstate its influence!

That Apple is an industry giant today says nothing about their importance in those early days. The Apple I was simply not very (at all?) influential. It's only important now in that it was the first product of a company that would later become very influential in the industry. The revolution, however, was sparked by the Altair -- the Apple I was hardly a contributor. It's real a stretch to say that that particular machine had *any* influence at all.

Re:Apple I?

[narcc]

Forgot to mention this bit.

The Apple I was much closer to a modern all-in-one system,

Not really. Not even close. Aside from the fact that it was less complete than other offerings at the time, even purchased assembled (assuming any of the 200 were sold that way) There were already products on the market that were MUCH closer to a modern system like the AIM-65 that had, at purchase, an integrated display, keyboard, power-supply, etc. The Apple I was a bag of parts and a circuit board. It didn't even have a case option.

The Apple II is very likely the machine that you're thinking of, though they were just another me-too player in 1977. Apple would have had a helluva time if it weren't for VisiCalc. Had that been produced initially for a different system, computer history would look very different. VisiCalc drove the adoption of the personal computer in business and, as a consequence, drove sales of the Apple II.

The TRS-80 Model 1 (the worlds first mass-produced personal computer) sold for less than half the Apple II in 1977. Had Bricklin written VisiCalc for the TRS-80 instead, Apple may have vanished like so many other early players.

the real advantage of the A2 was the availability of lower cost, higher capacity, easier to use disk drives, [...] That in turn made Visicalc practical. Even though it was first on the Apple ][, Visicalc simply didn't work as well on a PET or TRS-80.

A strong argument. Though you have to admit that had it been created for a different computer first, it would have advantaged Commodore or Tandy over Apple in that rush to adoption. Woz's disk drive was brilliant, I won't argue that. However, I don't believe that having that edge at the time would have moved Apple's over cheaper alternatives. At the time, if you wanted VisiCalc, you bought an Apple not because it ran the software better, but because Apple became associated with that program. Had VisiCalc not been produced for the Apple II at all, I don't know that they'd have survived into the 1980's.

Re:Apple I?

[Anonymous Coward]

Because the USRP is already the equivalent of an Altair?

Re:Apple I?

[msauve]

No, the a2 is not the machine I'm thinking of. The AIM had a 1 line alpha-numeric "display." It did not include a case, although they would sell you one. It did not have a power supply, although they would sell you the 4 voltage regulated one it required along with an expensive metal case. The Apple I had integrated 40x24 video. The Apple I was not "a bag of parts," it came fully assembled, other than keyboard, TV/monitor, and a couple of transformers. The Altair/IMSAI/Poly/etc. needed an expensive add-on I/O card, and were used primarily with external terminals (ASR-33, I/O Selectric, or ADM-3a, only $900+), although you could get video out for a couple hundred $$. If you want to point to a S-100 system with influence, it would have to be the SOL-20, which was a nicely packaged all-in-one system, with keyboard and video, but it was $1500 assembled.

Re:Apple I?

[Nethead]

Hey! Don't leave the SWTPC 6800 out of that list.

Re:Apple I?

[narcc]

You're confusing "assembled" with "influential". That doesn't even make sense.

As for the ridiculous criticism of the AIM-65's display, remember that it was not uncommon in to the 1980's, at least on portable computers. Compared to the popular SIM and KIM units of the time, the AIM-65's display was top-notch. It was also FAR more "complete" a system (why this makes you think it's influential, I'll never know) than the Apple I.

On the Apple I, you forget your history. About 200 were made and were not all sold assembled nor sold with a display, case, keyboard, power supply, etc. It was very much a bag of parts. I'm not sure you could even buy it pre-assembled (unlike other kit computers at the time), I'd have to dig around a bit to find out.

As for it's influence, I can't find any. The Apple I at the time was just another no-name kit computer with a tiny production run. Hardly any were produced, and it certainly didn't set any trends in the industry or spark a revolution and pioneer an entire industry like the Altair 8800.

Why am I even arguing this? Go read a damn history book. The only reason anyone gives a shit about the Apple I is Apples place in computer history, not because of that particular unit! As you seem to think that the Altair wasn't influential at all, you CLEARLY need a history lesson!

Re:Apple I?

[msauve]

"As for it's influence, I can't find any."

Your problem, not mine. Ref: Kilobaud #2.

Re:Apple I?

[narcc]

From a guy who thinks the Altair 8800 hardly had any influence, I'm not surprised that you've confused "Apple I" with "SOL-20".

Re:Apple I?

[msauve]

For a guy who expounds COBOL, you're consistent.

Re:Apple I?

[narcc]

Indeed, just like COBOL. It runs the world, you know. :)

A cheaper solution

[Anonymous Coward]

Use this device for $25, covers 65-1700Mhz, 8 bit and 2.8Msps:

http://www.ebay.co.uk/itm/DVB-T-USB-Stick-RTL2832U-Elonics-E4000-Receiver-RTL2832-RTL-SDR-Digital-TV-/150809364547

Then use either GNURadio, or HDSDR to view the data and listen to radio, TV, etc.

Beats the hell out of an $850 product that is limited to one desktop machine.

Obnoxious Apple Namedropping

[Samantha Wright]

The TRS-80, the SOL-20, and the PET 2001 were also officially introduced in 1976. (In fact, the SOL-20 dates to '75... as does the freaking Altair 8800.) I'm pretty sure the TRS-80 was more popular than the Apple I and hence had more direct impact. Ars, you sadden me this day for ignoring these other systems.

Re:Obnoxious Apple Namedropping

[Man On Pink Corner]

And the time machine was officially introduced in 2012, apparently, if your post is to believed.

Re:Obnoxious Apple Namedropping

[Samantha Wright]

Ugh. That's a bad header. Okay, so the TRS-80 and PET were a year later; I blame the weird formatting on that page. Still—the Altair was already out. That deserves way more respect than the Apple I.

Re:Obnoxious Apple Namedropping

[Man On Pink Corner]

I don't agree. The Altair was more Hermes than Prometheus. It brought the message down the mountain, but it didn't bring the fire.

As delivered, the Altair was programmed with switches, not an ASCII keyboard. It had no video output capability. No sound, no graphics, not even a rudimentary text display unless you duct-taped it to a Teletype machine. Some of these shortcomings were later remedied with optional peripherals, but still, it was nothing like an Apple I, much less an Apple ][.

The early Apple and TRS-80 machines were much more influential, and much more recognizable today.

Re:Obnoxious Apple Namedropping

[Samantha Wright]

Prometheus the Apple 1 is not. There were only 200, after all; the TRS-80 Model I sold ten thousand units in its first month and a half of sales. The Apple ][, sure, but if you consider not shipping with a keyboard to be a critical threshold in a microcomputer's ability to support the masses, then surely the absence of a significant user base is more important than coming out first.

There are much cheaper SDRs

[Anonymous Coward]

Try rtlsdr.com in a few days after they recover or websdr.org.

I'm building a Softrock Ensemble II $70 from kb9yig.com

The UPSDR and Phi are far more capable, but at a price.

Re:so? old hat.

[Anonymous Coward]

I think the cool thing about this SDR is that this is a transceiver. They don't give many details, but if they've gotten a true software defined transmitter, I'd say that this is a pretty big achievement.

Speaking of which, does anyone have any details about the power output of this thing? The website doesn't give many details.

GQRX

[SealBeater]

Hopefully this guy won't be mad at the shoutout.

There is a lot of work being done to make GnuRadio in general more accessable

GQRX http://www.oz9aec.net/index.php/gnu-radio/gqrx-sdr

Re:GQRX

[Man On Pink Corner]

There is a lot of work being done to make GnuRadio in general more accessable

If they wanted to do that, they could start by shipping Windows binaries that would work with the Funcube and other dongles.

Re:GQRX

[thygate]

there's also HDSDR (http://www.hdsdr.de/screenshots.html), and sdrsharp (http://sdrsharp.com/index.php/downloads) for windows.

Re:GQRX

[thygate]

GnuRadio is not a front-end, it's a framework (mainly python and c) for processing signals (DSP) with lots of functions for filtering/demodulating/.. There is gnuradio-companion which allows you to design in a GUI where you place blocks and connect them.

Re:GQRX

[Anonymous Coward]

There is a lot of work being done to make GnuRadio in general more accessable

If they wanted to do that, they could start by shipping Windows binaries that would work with the Funcube and other dongles.

GP said "more accessible", not "accessible to retards with toy computers". Give it another couple of years...

Re:GQRX

[Anonymous Coward]

This is why you lose. That, and the whole not-bathing thing.

Re:gnu radio

[englishknnigits]

People like bad mouthing Stallman (and, unfortunately, by association GNU/FSF) because he is an arrogant, condescending, overly confrontational, sexist, hygienically inept, eye sore of a man (I'm sure I missed about 10 more glaringly obvious, negative attributes). That doesn't make him wrong or FSF a bad thing but it does make him an easy target for bad mouthing. He is the type of man you want in the background of a movement, locked away in a server room, trying to push the movement forward. He is not the type of man you want to be front facing for any cause. It is just asking for ridicule and derision (deserved or not).

Re:so? old hat.

[AmiMoJo]

SDR isn't software controlled, it uses software to define the radio protocol in use. Could be as simple as AM or FM modulation, FSK, Manchester or some really complex frequency hopping madness. The point is that a single highly flexible receiver is connected to a DSP that can then replace any number of specialist radios.

Mobile phones already use it. A DSP can process various network protocols like GSM, CDMA and LTE which in the past would have had dedicated decoding/encoding circuitry for each.

Not a full solution

[nurb432]

It doesn't transmit.

Freq-Hopping Encrypted Tac-COMM

[BlueStrat]

One thing to note that I haven't seen mentioned yet is that this goes a long way towards making secure & encrypted tactical radio communications much, much more do-able and affordable for private citizens. A capability that's up till now largely been restricted to LEAs and the military.

This unit's flexibility make setting up frequency/band-hopping and encryption relatively easy. This capability in civilian hands is sure to be disliked by US TLAs and police.

It makes me wonder whether the government will attempt to outlaw certain programs and/or regulate what software is "legal" to have loaded in such a device, and/or require device capabilities be hardware-crippled/restricted to be legally sold.

After all, according to the government, it's right and proper that the government conceal it's communications and activities from the citizens, but citizens may certainly not be allowed to communicate securely without the government being able to monitor if they wish.

"The liberties of a people never were, nor ever will be, secure, when the transactions of their rulers may be concealed from them." - Patrick Henry

Strat

Re:Freq-Hopping Encrypted Tac-COMM

[Anonymous Coward]

Since it requires strapping a workstation-class PC to your back to run the SDR, which is just a teeny tiny bit non-tactical, I doubt the law enforcement agencies really give a shit, tinfoil hat boy.

Re:Freq-Hopping Encrypted Tac-COMM

[BlueStrat]

Since it requires strapping a workstation-class PC to your back to run the SDR, which is just a teeny tiny bit non-tactical, I doubt the law enforcement agencies really give a shit, tinfoil hat boy.

Says who, anonymous-coward boy?

Only if you plan on doing development work while at the protest/demonstration.

Once you've got a fixed task like frequency-hopping then it's simply a matter of providing just enough computing power to keep the SDR on-task. A Raspberry Pi would likely be overkill, and could likely handle both the frequency-hopping and encryption.

Strat

Re:Freq-Hopping Encrypted Tac-COMM

[Anonymous Coward]

>A Raspberry Pi would likely be overkill, and could likely handle both the frequency-hopping and encryption.

How sad; you don't even understand what a SDR is and how it differs from a conventional radio. Oh well.

Re:Freq-Hopping Encrypted Tac-COMM

[BlueStrat]

>A Raspberry Pi would likely be overkill, and could likely handle both the frequency-hopping and encryption.

How sad; you don't even understand what a SDR is and how it differs from a conventional radio. Oh well.

How sad; *you* don't understand that a hardware card just wants certain data and voltages on certain pins to perform a certain task and doesn't care if a Raspberry Pi, Big Blue, or an infinite number of monkeys throwing switches on the head of a pin provides it. It's just the card's chips handshaking with your hardware's chips and transferring data and instructions.

I guarantee you that the test jig at the end of this card's factory production line is *not* a full PC running gnuradio, but a logic board with PROMs loaded with the necessary data commands and the I/O hardware to put the card through it's paces. You simply don't need all the fancy GUI software and PC OS to have a hardware card like that perform certain limited, fixed functions. Especially since this card's full specs will be open.

Strat

Re:Freq-Hopping Encrypted Tac-COMM

[makomk]

This thing connects to the host computer - which does the brunt of the signal processing - over a "8Gbps, low latency, PCIe x4 bus". The Raspberry Pi does not have a PCIe connector, or anything with close to the bandwidth of even a a PCIe x1 interface. That's before you even get to the processing power required...

Re:Freq-Hopping Encrypted Tac-COMM

[shiftless]

Then perhaps you could explain it, asshole? I'm a satellite communications engineer and his idea sounds workable to me. I don't know if a Raspberry Pi would have enough computing power, but some kind of low power VIA system or other embeddable X86 system would certainly do the job. All the "modulation"/signal generation is handled by software, and frequency upconversion by the hardware, right? (I didn't read the link yet.) Then all you need is what, a power amp and antenna? Where's the problem?

Re:Freq-Hopping Encrypted Tac-COMM

[Anonymous Coward]

Because modulation and signal processing isn't computationally trivial. Unless the task you have in mind is so insanely trivial that you can do it all with LUTs on the software side, without any real-time mathematics, you need a beefy CPU.

Emulating what used to be handled by $0.35 analog componentry in software is not computationally trivial. Those sorts of DSP tasks have traditionally been done in DSP-specific CPUs and FPGAs and ASICs. The "coolness" of SDR is that desktop-class computers have become fast enough to do high-speed tasks that have
traditionally been done elsewhere.

SDR is distinct from "Software Controlled Radio" -- whose host-side computational requirements are indeed quite modest, since there's no DSP involved.

Re:Freq-Hopping Encrypted Tac-COMM

[Anonymous Coward]

I'll just open up my Visual Basic GUI with a big button labeled "Start" then... The amount of connections doesn't matter, at the end of the day its just two chips shaking hands and communicating back and forth.

It's an FCC regulation, alright....

[Ellis D. Tripp]

[url]http://www.gpo.gov/fdsys/pkg/CFR-2010-title47-vol1/xml/CFR-2010-title47-vol1-sec15-121.xml[/url]

Re:gnu radio

[Anonymous Coward]

So he has Assburger's?

$15 DVB-T tuner (RTLSDR)

[thygate]

I'm surprised no-one has mentioned the RTLSDR yet. A $15 DVB-T Tuner than can tune from ~70MHz to 1700MHz. Maximum bandwidth is about 2MHz. It has poor dynamic range (8 bit ADC), but for receiving strong signals it's awesome. There is a source block for gnuradio, and some nice tuners for windows (HDSDR, sdrsharp, ..). Lots of cool stuff to do. For instance I've successfully received MODE-S transponder replies from airplanes as far away as 200km with the stock antenna. Tuning to FM radio, portable mobile radios, DECT, GSM, Exploring the spectrum, .. Of course it cannot compare to an USRP or this new Phi, but it's very cheap and is perfect for getting started, and does not require a HAM license. check here : http://sdr.osmocom.org/trac/wiki/rtl-sdr

Re:gnu radio

[lister king of smeg]

(I'm sure I missed about 10 more glaringly obvious, negative attributes).

you forgot; he is fine with pedophilia and approves of necrophilia, and bestiality, insest, prostitution, just to name a few, he is paranoid about the use of cellphones as tracking devices and espionage tools, oh and he incredibly jealous of tordvalas and linux.

Re:so? old hat.

[JumboMessiah]

Yes, there are many HAMs around working on custom SDRs. HPSDR is one I have some exposure to. It handles RX/TX and comes with open schematics. There are some HAMs doing some really cool stuff with it.

Recieving is restricted

[jrincayc]

Receiving is restricted as well. Ellis D. Tripp (755736) also posted on this:
http://www.gpo.gov/fdsys/pkg/CFR-2010-title47-vol1/xml/CFR-2010-title47-vol1-sec15-121.xml

"scanning receivers and frequency converters designed or marketed for use with scanning receivers, shall ... Be incapable of operating (tuning), or readily being altered by the user to operate, within the frequency bands allocated to the Cellular Radiotelephone Service "

See also http://en.wikipedia.org/wiki/Advanced_Mobile_Phone_System#Frequency_bands

test equipment exception

[jrincayc]

In order for the test equipment to apply they must be "marketed exclusively as test equipment" title 47 vol 1 15.121(c). However from the website http://pervices.com/about.html it states "Phi can capture over the air waves, so with the right app, you can watch cable for free." Therefore, Per Vices is marketing the Phi to areas besides test equipment users, so it is illegal.

Re:test equipment exception

[TwineLogic]

Test equipment is defined in 47 CFR 15.3 (dd). However, your reference to 47 CFR 15.121(c) is helpful -- it indicates that "scanning receiver" laws may be surmountable. The SDR regulations seem to still apply. I realize the intention may have been to cover transmitters, but the regulation covers all Software Defined "Radios."

By the way, "Title 47" doesn't disambiguate 47 CFR (regulations, written by the FCC) from 47 USC (laws, written by Congress).

And ITAR rears its ugly head

[Anonymous Coward]

more to the point.. selling such a device will put you squarely into export control territory. spread spectrum radios, especially if designed for anti-jam/low-observable, are munitions. Better make sure the people you sell,give, transfer etc. the software to are US persons and agree to the export controlled-ness.

WinRadio And Yes, It Runs Linux

[westlake]

I seem to remember a win95 radio card that slid into an AT slot back in the mid or late 90s...

WinRadio is still very much alive.

WinRadio builds SDR sets for marine, advanced hobbyist, and professional applications. Expect to pay $900-$1000 at entry level.

The WiNRADiO WR-G39DDCi 'EXCELSIOR' is a high-performance HF/VHF/UHF/SHF software-defined receiver with a frequency range from 9 kHz to 3500 MHz, with two independent channels of 4 MHz wide instantaneous bandwidth available for recording and further digital processing, plus a 16 MHz wide real-time spectrum analyzer.

WinRadio

The truth is finally out, fuckers!

[Anonymous Coward]

Apple invented radio. So there.