Slashdot Mirror
First Fully Digital Radio Transmitter Built Purely From Microprocessor Tech
Zothecula writes For the first time in history, a prototype radio has been created that is claimed to be completely digital, generating high-frequency radio waves purely through the use of integrated circuits and a set of patented algorithms without using conventional analog radio circuits in any way whatsoever. This breakthrough technology promises to vastly improve the wireless communications capabilities of everything from 5G mobile technology to the multitude devices aimed at supporting the Internet of Things (IoT).
No actual info in article, just hype and buzzwords.
Strange things are afoot at the Circle-K.
Must be a slow news day. You can't patent algorithms
Harrison's Postulate - "For every action there is an equal and opposite criticism"
This. Or This.
Sounds way too good to be true... but it should make the lives of the FCC a living Hell... I mean how many people would willingly use the crowded 2.4 GHz spectrum for anything if they had a cheap and easy alternative? Looking forward to apps named FreqManip for all of my bottleneck bypassing needs... yeah I'll believe it when I see it.
No problem generating RF with digital circuits; it has been done for decades. The trick is to generate clean frequencies with no significant harmonic content and no spurs. Now perhaps the cicuits used in this article are digital and analog on one silicon substrate. Certainly a DAC can be made using a semiconductor resistor network fabbed on the same substrate as the digital electronics. And capacitors and inductors can be fabbed on a silicon substrate. I would like to see the details of what they are actually doing. "Pure digital" sounds like a phony to me.
OK, no real technical data and some absurd claims here.
First all-digital transceiver? No. There have been others. Especially if you allow them to have a DAC and an ADC and no other components in the analog domain, but even without that, there are lots of IoT-class radios with direct-to-digital detectors and digital outputs directly to the antenna. You might have one in your car remote (mine is two-way).
And they have to use patented algorithms? Everybody else can get along with well-known technology old enough that any applicable patents are long expired.
It would be nicer if there was some information about what they are actually doing. If they really have patented it, there's no reason to hold back.
Bruce Perens.
The article seems to be full of PR, but from what it describes I'm guessing this is a delta sigma front end and they are selecting off a suitably placed alias to do the modulation. If they have managed to get the noise shaping right then conceivably a standard antenna could suffice as the aliasing filter, which would be quite an achievement. Also getting the timing and jitter performance right are tricky issues they would have to be solved in a low cost product. I can't imagine what else they could mean by all digital, and if they've built this then it is actually quite an exciting development for low power radio. Definitely could bring the cost of systems down by a lot.
People have done this on TI calculators (& likely other systems with similarly little shielding & sufficient clock rates). No hardware support needed—just cause some long enough trace (e.g. on the data bus) to oscillate at the correct frequency. Granted, a 6 MHz Z80 can pretty much only only do AM radio (& can only be picked up right next to the radio), but the principle is not new.
Is built from analog parts.
where's the beef, Dice?
I could build a "Fully Digital Radio Transmitter" in a few minutes using a Crystal and a CMOS gate.
The 1620 was an all digital machine built of discrete transistors. As an undergraduate, we wrote programs that caused the machine to alternate between two loops at a variable rate. The computer radiated so many harmonics that this could be heard all across the AM band where no strong station existed. We programmed it to play (mostly) classical music, "Flight of the Bumblebee" was the perennial favorite. Any truly all digital transmitter will generate harmonics outside of the allowable FCC band, so at the very least they need a really good analog bandpass filter on the output.
All digital circuitry is very noisy. It generates harmonics and switching noise over a huge range of the radio spectrum, meaning it MUST have output filtering to work in the real (legal) world. Analog circuitry uses less power for the same RF output, making it a poor choice for mobile use. And if it's not mobile then you won't care about adding a few capacitors and resistors to filter the output.
This is not a first. I did this a couple of years ago with an FPGA. Fully digital synthesis with a delta-simga DAC. I hung a 1M wire off the CMOS output, and got a range of about 50'. Pretty basic software radio stuff.
All digital transmit is easy and has been done with FSK, BPSK, OOK, and numerous other K's by many people lots of different ways. We do this as a class exercise at 228 MHz using a Digilent FPGA board (Nestor, J. A., & Nadovich, C. (2009). An FPGA-based wireless network capstone project. In 2009 IEEE International Conference on Microelectronic Systems Education, MSE 2009 (pp. 53–56).) Much more difficult is all digital _receive_ as you need to conquor the selectivity and sensitivity problems. It's hard to avoid using at least a balanced mixer in the receiver.
It's very easy to make a digital transmitter. It's very hard to meet the thousands of compliance specs (spectral emission mask, spectral flatness, adjacent channel interference, etc).
One day this will happen, but I'm not convinced it's today.
A little late. We did it with AM transmission using traces on PC motherboard years ago. Just had to send data patterns to generate RF waveform in AM band.
Don't forget the musical hard-drives we had as well.
http://hackaday.com/2014/06/15/easily-turn-your-raspberry-pi-into-an-fm-transmitter/
Basically they are bit-bashing FM on a GPIO pin
If motherboards came with this maybe your desktop pc could be used for simple multiroom audio
"The significance of this new technology cannot be overstated:"
I believe you just did.
How is this different than the Intel Rosepoint chip?
http://venturebeat.com/2012/09/13/after-a-decade-of-research-intel-shows-off-its-digital-radio-chip/
Uh, The spark gap transmitters used by Hertz and Marconi were digital (Morse code is a digital protocol) and for the most part the only tuning was done by the antenna. Latter there was some sort of tank circuit or resonant tuning added, but I don't think so in the beginning.
Raspberry Pi has software that will do it, up to a few hundred MHz
Too bad its patented, means it will newer take over the world...
I already listen to music from my Raspberry Pi using it as a transmitter with no analogue components - http://naich.net/wordpress/?p=...
Depending on the frequency band (i.e. not terahertz), it's not that hard to make an arbitrary digital signal encoder that produces an analog signal, after a little bit of filtering. For instance, it's common enough to do audio by wigging a single bit and then passing that through a low-pass filter.
If they're claiming to do it without ANY analog hardware, then I call BS. However, there is filtering inherent in the digital circuits. I could imagine using a genetic programming to learn code sequences that produced desired signals as basically cross talk. However, process variation could really muck with that.
As for receiving, that's not too hard either. Say you want to receive a wide band of frequencies. You can make a band-pass filter in the analog and convert that to digital using an ADC. Once in the digital domain, you can separate the channels using fourier analysis or Taylor series.
There have been a lot of people (too many references to site) who do things like make FM transmitters out of a Raspberry PI.
I used to play with making my Apple ][+ do wonky things to transmit weird noise to my FM radio back in the day...
From GTA V.
Digital is just an interpretation.
The press release is pretty ridiculous on many counts. Been thought of and done long ago just being a start.
They have been doing this with sound for some time. Radio is just faster. (Yes, I know that is WAY oversimplified). At radio frequencies, any electrical engineer will tell you there is no such thing as digital. The edge of a square wave is not perfectly straight. It is a noisy curve based on the impedance of the circuit and the current used to drive the transition. There is inductance and capacitance in every conveyance of electricity. In a "clean" circuit, the effect of this parasitic L/C is either negligible or compensated for.
A radio antenna is, by definition, an analog part which electrically resembles a coil with some capacitance . So even the title misleading. The fact that they can us algorithms to control the digital signal in such a way that the antenna will smooth it out into a radio wave is kind of cool, but it isn't a crazy breakthrough. You can see almost every computer on a spectrum analyzer as radio wave source. This is just a neat trick, like getting the line printer to sound like music by sending the right stuff to print.
This isn't new at all, far less "the first". The Altair 8800 didn't have any sound capabilities so people used to tune there radios to the clock frequency of the main data bus and feed data to it at frequency's that were modulated by the bus speed. It sounded like crap but it was far from a tuned system and that was back in the 70's. https://www.youtube.com/watch?... I might believe there is some type of breakthrough in this if they had actually put any real information about it in the article but all that's in there is fluff.
Do I own royalties for turning my PI into a radio?
http://www.instructables.com/id/Raspberry-Pi-Radio-Transmitter/
http://hackaday.com/2012/01/26/sprite_tms-three-component-fm-transmitter/