Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

HackRF, an open source SDR platform

kickstarter.com

Post reply on HN
Print view View original

41–50 of 57 posts

Re: HackRF, an open source SDR platform

#41

Post by cdjk »

yyao wrote: post #33
gdc wrote: post #26

Yeah, the lack of < 30 Mhz means this would be a pass for me.

Depends on what you want to do. A lot of people think that they need more bandwidth or power to achieve something. This may not necessarily be the case. QRP guys have crossed continents on less than a watt. I remember from my undergrad days that we've played with band passing the human voice. We managed to get surprisingly good results with 500Hz of bandwidth. There's a lot of interesting things you can do with more…

Most QRP work is at frequencies less than 30MHz. While 6m might work, tropospheric ducting, sporadic E, etc are touchy enough that adding in QRP operation wouldn't be pleasant. The RF upconverter would help with that, however.

Power is easily solved by an external amplifier.

Increased bandwidth would be nice, but 20MHz is plenty. While you might need that for microwave experiments (wifi, gsm, etc, and spread-spectrum experiments), the big advantage of the bandwidth is being able to listen in to everything at once, and see where signals are at a glance, without tuning through frequencies. Being able to glance at an entire band, and immediately see where QSOs are happening is pretty cool. Or you could record all of the FM broadcast spectrum, and decode it at your leisure later. That's not too practical, but is pretty cool. And 20Mhz is wider than every ham band 2m and lower.

And 500 MHz isn't quite enough for voice - SSB is generally 2.3 - 3 kHz. 300-500 Hz is ideal for CW (morse code), and digital modes can use even less.

Re: HackRF, an open source SDR platform

#42

Post by ajb »

rsync wrote: post #21
est wrote: post #2

The spec from HackRF github wiki: https://github.com/mossmann/hackrf/wiki/Jawbreaker * half-duplex transceiver * operating freq: 30 MHz to 6 GHz * supported sample rates: 8 Msps to 20 Msps (quadrature) * resolution: 8 bits * interface: High Speed USB 2.0 (with USB Micro-B connector) * power supply: USB bus power

I'm interested in comparing this to the bladerf device: http://www.kickstarter.com/projects/1085541682/bladerf-usb-3... ... since I already have one. HackRF appears to be about half the size, which is very nice. Freq range is 30-6000, vs. the bladerf which is 300-3800. Bladerf is USB 3.0, however, and appears to have a higher sample rate ? "capable of capturing 40MHz 12-bit full duplex quadrature samples in realtime.…

The HackRF is half-duplex (can only either rx or tx at a given time - unless you get two). The bladeRf is full duplex.

Re: HackRF, an open source SDR platform

#44

Post by yyao »

cdjk wrote: post #41
yyao wrote: post #33

Earlier quoted context omitted.

Depends on what you want to do. A lot of people think that they need more bandwidth or power to achieve something. This may not necessarily be the case. QRP guys have crossed continents on less than a watt. I remember from my undergrad days that we've played with band passing the human voice. We managed to get surprisingly good results with 500Hz of bandwidth. There's a lot of interesting things you can do with more…

Most QRP work is at frequencies less than 30MHz. While 6m might work, tropospheric ducting, sporadic E, etc are touchy enough that adding in QRP operation wouldn't be pleasant. The RF upconverter would help with that, however. Power is easily solved by an external amplifier. Increased bandwidth would be nice, but 20MHz is plenty. While you might need that for microwave experiments (wifi, gsm, etc, and spread-spectrum…

> 500 MHz isn't quite enough for voice

Thats what I thought. We had a few teams look at ways to carry human voice with less BW in a sensory communications course. The winning team did it with 500Hz. I remember it clearly. There was a stunned murmur during their project presentation when they announced their results.

I don't remember the exact numbers, but the gist of it was, they applied 2 bandpass filters, one on a lower frequency range and the other on a higher frequency range. Both passbands were 250Hz wide. They combined the two signals together as a 500Hz signal. The pass bands were seperated by 500Hz or 1kHz IIRC. So pretty much they kept the lower and higher frequencies and threw away the rest. The result was pretty good. Human languages have a lot of redundancy in information and the quality was good enough for general conversation. It was comparable to what you have on a cell phone.

Re: HackRF, an open source SDR platform

#45

Post by patrickyeon »

nanospider wrote: post #40
xradionut wrote: post #23

Earlier quoted context omitted.

That's the nature of the RF business. Most folks are interested in one ( HF), or have money for equipment that handles both. Building good equipment that handles the full spectrum is challenging.

http://thinkrf.com/wsa5000/

Building that one was challenging too (I used to do RF work for thinkRF). Going back through history, they started off with something that would only cover a specific band (say wifi, or specific cell bands). Then we did the 0440 (400MHz-4GHz), then the 0108 (10MHz-8GHz). There are all kinds of difficulties that come in to making and RF front-end that'll cover all of that bandwidth (we also did amplification and filtering to get better quality reception).

You'll also note the prices, and that those are receive-only devices. Some of the reason the price is so high is that we did all the RF design in-house, designing to our specs. A lot of the cheaper boards out there have less amplification, no (or too little) filtering, and use an integrated LO/mixer/digitizer/baseband chip, which limits them to the specs that chip can handle.

Re: HackRF, an open source SDR platform

#46

Post by patrickyeon »

sparky wrote: post #18

Man it'd be nice if this (or BladeRF) went down to DC like the USRP, so you could also use it as an oscilloscope. It's easier on the USRP because of its motherboard/daughterboard architecture, which adds some cost and complexity, but the cost adder should be pretty small relative to the several-hundred-dollars these things cost.

The reason a fair number of these don't go down to DC is that they have an integrated mixer and ADC/DAC, so they are limited to what that one chip will do for front-end bandwidth.

Re: HackRF, an open source SDR platform

#47

Post by VLM »

aray wrote: post #32
oldgregg wrote: post #27

What would be the typical range for something like this or BladeRF?

I'd assume it depends heavily on your antenna configuration, line-of-sight, polarization, noise, etc. Edit: I sent a question on kickstarter asking about TX power limitations so if mossman answers it, the answer will appear in the FAQ section.

https://github.com/mossmann/hackrf/wiki/Jawbreaker

What it really depends on is your modulation method.

So say you want to operate legally as a ham radio op on 5760.1 operating CW. Well with a reasonably high gain antenna -15 dBm is OK. COTS MMIC amps and some filters and you'll have a vaguely competitive contest rover. I would imagine QEX/QST and the like will have ham radio band amp designs for a couple bucks soon enough after this thing rolls out. Looking at the cost of a FT-817 and a transverter, I imagine you could save quite a bit of dough using this device... Given $200 I could probably scavenge up some TR relays good for 5 cm and maybe a whole watt or so. Placed in front of a modest rover capable dish this could pretty much kick butt. Plus or minus phase noise, which hasn't been discussed, and frequency stability (I'm guessing no 10 MHz GPSDO input) (edited to add, and I haven't seen a noise figure spec yet either...)

On the other hand if you insist on trying to do NTSC ATV ham radio operation around 427.250 you'll find that a smokin +5 dBm on a 6 MHz wide signal will get you perhaps across the room.

Re: HackRF, an open source SDR platform

#48

Post by VLM »

alokm wrote: post #4

Are there any existing reliable alternatives?

For real usable ham radio SDR there's always

http://fivedash.com/

aka the former kb9yig operation.

I have a liteII combined kit built for 30 meters. Works pretty well! Wide enough BW that with a 192K sound card (or was it 96K?) I would listen to one of the aircraft weather transmitters.

Note this is narrowband low frequency rx and tx gear, whereas there's two separate "cultures" of SDR people and one lives to demod wifi and the other lives to do advanced DSP on 160 meters or whatever. And the two don't often meet.

Re: HackRF, an open source SDR platform

#49

Post by VLM »

SubZero wrote: post #5

Would this be able to say...oh, idk...spoof and overpower digital TV channels in a localized area? I need to know...for science.

Yes and if you run the numbers for a couple dBm out and spread across a 6 MHz ATSC signal, compared to the actual stations (depending where you live, I guess) you'll have a range of perhaps feet. Maybe tens of feet.

Shannon's law is not to be denied! (And what decade did it rename from "Shannons Law" to "Shannon–Hartley theorem" anyway?)

Re: HackRF, an open source SDR platform

#50

Post by VLM »

benjamincburns wrote: post #34
cottonseed wrote: post #30

This is great. I met briefly with a company developing software radio products that led me to believe such devices are illegal or require special licensing. Is my impression wrong or has this changed recently? Second, can this do GSM, receive at least? Frequency band works, but maybe it doesn't have the bandwidth?

It's been a while since I've looked into this, but last I knew it was legal to listen to any band you want, but illegal to transmit on any band for which you don't posses the required license, if any. Then there's all kinds of rules about transmit power. For instance, last I knew you could transmit unlicensed on the public FM and AM radio bands so long as you're under some minimal threshold - I think it was 1W.

"I think it was 1W."

LOL no, you know .gov they make it as complicated as possible. If I recall the AM is DC input power limited and the FM limits are some weird field strength which is non-trivial to measure. Google for 47 CFR 15 and you're looking for part C 15.221 and .239 probably. You can read this stuff for free at gpo.gov.

There are about a zillion other part 15 bands you can legally operate under if you're willing to bother figuring out how.

If you're smart enough to figure out how to legally operate part 15, you're certainly smart enough to get a license to operate part 97, which doesn't have even a fraction of those goofy restrictions and is the original techie social network.

Post reply on HN