LimeSDR | Hacker News Reader

LimeSDR

limemicro.com

122 pointsoctopoc126 comments

I wish them luck, but judging by the ~week I spent trying to get a LimeSDR mini to work, their product is so half baked that it's basically not fit for purpose, and if you're considering getting one, I'd recommend something with an Analog Devices front-end instead.

The PLL would fail to lock at random frequencies and temperatures. You'd try to scroll through e.g. the 2.4 GHz ISM band and would randomly get downconverted signal or baseband noise on every other click. Two minutes later, the set of bands you could capture would be entirely different. There was no consistency whatsoever.

When I went to their forums to see if other people had encountered the same issue, it turned out that basically none of the LimeSDRs even pass their own self tests [1]. I don't know how they're still selling these devices... they don't work.

I returned it, paid ~4x more for a USRP, and could not be happier. The hobby is playing with radio modulation schemes, not debugging a defective piece of silicon.

[1] https://discourse.myriadrf.org/t/new-limesdr-mini-loopback-t...

Unfortunately, I agree with this. I was an early backer on the mini, and got it with the aluminium case. While the case looked great, it mechanically interfered with the antennae, meaning they didn't actually make a connection. Their support taught me how to screw in an antenna, ignoring the actual problem, before deciding my issue was too hard and ignoring me. I got eventually got in some antennae from Ossmann, but when other hardware/software issues cropped up, my enthusiasm for the device had already waned too much to want to put in effort. I got some basic things working eventually, but never outperformed cheaper devices.

I've got the LimeSDR standard size one, and I've never been able to make good use of it. Thinking of throwing it on eBay soon.

Software support is far more limited than I'd prefer compared to other SDRs too, which someone at DEF CON told me was because they pissed off the community and copied of a lot of things from another SDR project. I don't know how true it is, but I can see that support on all sides is pretty half baked.

Argh, I'm jealous.

I've been looking for suitable SDRs in the ~£180 range for a few months now, and nowhere has one suitable for 5G NR or 4G LTE available. HackRF seems to be pretty crippled in mobile networking, and Pluto seems to not work at all on srsRAN/srsLTE.

I doubt you'd be willing to let go of your full-size LimeSDR for £180, though, and I don't blame you based on its demand level.

I bought LimeSDR Mini back at crowdfunding campaign and didn't have good experience too. Janky middleware, problems with tuning, randomly occuring errors. I haven't used it in quite a while and I hoped these are all software problems they would fix eventually, sad to hear it is still broken.

Which USRP do you use? And what do you do specifically with it?

A B205.

Mostly playing around with high spectral efficiency digital modulation techniques like PSK/QAM and OFDM using GNU Radio. Trying to shovel as much data as I can between two links with as little power as possible. I'd like to experiment with below-the-noise-floor communication, but haven't gotten there yet.

It's sort of silly, because you can do all of this in simulation, but it's more fun with real hardware. Eventually I'd like to be able to build a (solar?) drone with a radio system resembling a cell tower and do ground-drone-ground relay experiments, but I'm extremely not there yet.

I also wanted something that could transmit after I got a ham radio license (out of boredom during the pandemic, shortly after they started doing testing online), but haven't really used it that much, as I don't really have a good antenna set up. For data experiments, a wire or a PCB trace works well enough.

I also do a lot of embedded Wi-Fi / Bluetooth stuff at work and would like to use this for packet captures, but haven't needed to yet. It's completely overkill for something like that though.

Wait, they're doing the exams online now? My laziness to go to a testing center has been holding me back this whole time

That's what I said... :-)

I used hamstudy.org and took a test administered by GLAARG over Zoom.

https://glaarg.org/

Very much this... two good combos, although the laurel VEC is also fantastic.

Get your ticket, its a fantastic experimenters hobby... just ignore the stodgy old man part of it

As mentioned, GLAARG and others does them online. However I'd like to throw out that amateur exams are held all over the place, quite often. These are typically friendly events and is a good way to meet people that have been in the hobby for a long time. Typically, your local VE team is part of a radio club that can tell get you lots of good getting started tips, especially if it pertains to local activity.

https://www.laurelvec.com/?pg=exams (All Laurel VE sessions are free and processed next business days)

http://www.arrl.org/find-an-amateur-radio-license-exam-sessi...

You could take a look at Intel's Falcon 8 drones [0]. It might get you thinking differently, and the things they pull off a the Olympics is amazing.

[0] https://www.intel.com/content/www/us/en/technology-innovatio...

Yep, the USRP series is solid. National Instruments buying them was really smart.

I bought me one a year or so ago with some big project plans that I just haven’t started. Sadly my math is not up to par for doing SDR work or Signal processing.

This is actually a shame. I looked into getting one of the cheaper USRP's (I think the cheapest options run a little more than the PCIe limesdr)

I had the exact same experience. It’s junk. Get a USRP if you can, or perhsps a bladerf if you’re a hobbiest.

Tried to make it work for a big project. Spent a lot of time debugging. I was suspecting the code at first. But finally realized the Radio wasn't working correctly. Couldn't return it or get a replacement.

This has been my experience. I traded mine out for a HackRf

SDR is one of those technologies I still want to play with that I feel will open up this incredible new world to my curious mind. All of a sudden, the invisible waves around us that we often tangentially know about but never think about (because we can't see, smell, or hear them), become a whole sixth sense to explore.

I really hope for some free time to play with this, and to expose this invisible world to those around me (especially my kids), in a way that has been much less accessible in the past.

Even things like ceiling fan remotes, door sensors, wifi, a garage remote, a car key, ISS video feeds, data from stuff in SPACE, weather data, I mean, we're bombarded with it every day all the time and we just don't realize it.

Anyway, just had a wide-eyed moment, as I do every time I read about another thing that makes SDR just a bit easier to access.

The best one is the ADS-B Out from airplanes, you can map the planes around you with a $20 widget.

This sounds really cool. When you say 'around you', what kind of radius are we talking?

Hard to answer that it depends on so much; the quality of the rx you're using, the antenna and your local environment (open field? wooded area? valley? top of a hill?).

It really depends on the environment around you. I live deep in valley, so my range is pretty limited to line of sight.

You may be able to find out more information, including instructions on assembling your own ground station, at Flightaware - https://flightaware.com/adsb/piaware/build/

As others have said, it depends on terrain, your antenna, the altitude of the aircraft, etc. But generally, 50-60 miles is pretty easy with almost any setup.

I have an ADS-B antenna mounted on my roof that feeds into a filter and an amp, and I get less than 50 miles in some directions (because of mountains) and more than 150 miles in other directions (over the ocean).

Here's a map of my coverage (any altitude), with range rings every 50 miles: https://www.flickr.com/photos/wallofhair/51403416502

My coverage for altitudes up to 10,000 feet: https://www.flickr.com/photos/wallofhair/51404165241

It's radio, so nothing's guaranteed. Those maps show the maximum ranges at which I picked up any aircraft in that direction over a couple months, so they're kind of a best-case, showing when conditions possibly happened to be perfect. I also have a dead zone just around my house, because my gain is configured to prioritize range. Lots of tradeoffs in RF.

There's a site (with an impossible-to-remember name) that can predict antenna range for you (it's not obvious how to use it; Try "New panorama", then "Up in the Air"): http://www.heywhatsthat.com/

Here's what it shows for me. The blue ring is 40,000 feet altitude and the orange ring is 10,000 feet altitude: https://www.flickr.com/photos/wallofhair/51404944889

How do you get that footprint plot? I'm using dump1090-fa and don't see it

To answer my own question, in case others are interested: https://discussions.flightaware.com/t/how-do-you-create-a-co...

It depends on your location, antenna, and the aircraft's elevation, but in the bay area I'll often pick up aircraft all the way out to Sacramento. Sometimes quite a bit further.

Of course, the higher the aircraft the further you can "see" them. Most of the traffic I see is within 45 miles (especially low flying GA traffic), but I'll still pick up the occasional high altitude aircraft at 100+ miles out.

It’s very fun! I use it to monitor for low flying aircraft around a farm where I am doing some drone experiments. The legal drone limit in my area is 400 feet above ground level and near the hills some aircraft fly pretty low, so I’m going to ground the drone any time an aircraft is nearby that is not above 10kft.

I highly recommend not relying on SDR ADS-B reception alone and also checking online data systems for ADS-B that also use multilateration to track aircraft using their transponder when they aren’t ADS-B capable. Otherwise you might be operating under the assumption there are no contacts in your working airspace when there is aircraft present but not announcing their location.

https://flightaware.com/adsb/mlat/

You need both data sources—a number of planes are “delisted” from the commercial trackers and won’t show up, but they still broadcast ADS-B. You can find partial lists of such registrations through a keen Google search.

https://www.adsbexchange.com/ does not blacklist or delist aircraft tail numbers fwiw.

Sounds like a good idea. Thanks for the suggestion.

Hope it helps.

160 miles with a £6 rtlsdr and an antenna I bodged out of coax. Its painfully easy. Also have a nosy at POCSAG - there's still a little bit left in the UK.

Similarly you can map out lightning strikes with similar tools/methods.

Can you provide some resources about this?

There is the Blitzortung.org site - https://www.blitzortung.org/en/cover_your_area.php - which has some interesting information.

Thanks, but that link appears to be about dedicated modules. I already have a HackRF One and an RSPDuo, the person I was replying to seems to indicate it was possible to use a typical SDR.

If you want to play without spending too much money or time, yo always can get a chinese DVB-T USB receiver or similar with a RTL chip. It cost less than 50 bucks and usually works well without too much configuration AFAIK. Don't expect an incredible SNR or wideband, but it's pretty good for tinkering.

I got a LimeSDR when they first made them. Unfortunately the whole SDR software ecosystem is pretty terrible. Even with something as popular and widely supported as RTL-SDR most of the software is unusuable (with some happy exceptions). It also doesn't help that there is an absolute mountain of SDR software out there, but 99% of it is "this is cool, I'm going to write... eh nevermind" type stuff. There's so much ( https://www.rtl-sdr.com/big-list-rtl-sdr-supported-software/ ) that I can't even find the one good one I found last time I used an RTL-SDR.

I guess that is what happens if you play with software written by uber-nerds for uber-nerds.

Anyway in conclusion I would NOT recommend getting a LimeSDR unless you are really really into SDR and it's your only hobby, or better yet, your job.

as an early limesdr buyer i completely agree

The LimeSDR Kickstarter made these grand promises of running a repo full of usable apps for your board. Many years later that repo is still quite spartan.

I've generally had pretty good luck with my LimeSDR, but I don't try to push it too hard. The documentation can be frustratingly minimal in so many places and the USB implementation on my board is touchy (it must be plugged into exactly the right port on my computer or the device enumerates incorrectly and is unusable).

Some past related threads:

Investigating the Galileo Satellite Navigation System Outage with a LimeSDR - https://news.ycombinator.com/item?id=20464852 - July 2019 (41 comments)

LimeSDR Now Backed by the European Space Agency - https://news.ycombinator.com/item?id=16493727 - March 2018 (38 comments)

LimeSDR Mini – An open, full-duplex, USB stick radio for femtocells and more - https://news.ycombinator.com/item?id=15318233 - Sept 2017 (46 comments)

LimeSDR funded, some hours left to still get a discounted SDR - https://news.ycombinator.com/item?id=11946239 - June 2016 (10 comments)

LimeSDR GSM Base Station Demo - https://news.ycombinator.com/item?id=11931072 - June 2016 (27 comments)

IoT wireless hacking – LoRa on the LimeSDR - https://news.ycombinator.com/item?id=11882491 - June 2016 (1 comment)

Digital Video Transmission using LimeSDR and GNU Radio - https://news.ycombinator.com/item?id=11856912 - June 2016 (8 comments)

LimeSDR: Flexible, Next-Generation, Open Source Software Defined Radio - https://news.ycombinator.com/item?id=11610521 - May 2016 (28 comments)

SDR made me understand how much RF noise my desktop computer emits. Lots of spikes between 400-500mhz. Could be my monitor also, as scrolling text seems to emit the most noise.

Same here with the monitor, I have an old Dell 24" from 2006 that I somehow keep around. I've learned not to expect much if my ham radio is within 3 feet of it. Probably fixable with some clip-on parts but for now it's easier to scoot the chair over...

My Samsung monitor does that.

It's bad enough that I lose several channels on my Samsung TV when my Samsung monitor is on if the TV antenna is in the same room as the monitor.

interesting... I guess I never scrolled while looking at wideband noise :) I'm going to see if I can determine what is/was displayed on the screen by filtering/machine learning the noise, fingers crossed

There's a fun toy I just got called 'TinySA' (tiny spectrum analyzer). For $60, plus an adapter + some wire (another $20 maybe), you can get a live battery powered little receiver that shows noise across the whole spectrum from less than 1mhz through 350mhz.

A really basic loop of wire is a directional antenna, so you can narrow down exactly what is causing issues too.

I'm doing it to find noise issues for my ham radio setup, but it can be generally fun. Not quite an SDR, but to find noise sources it's pretty amazing for so cheap.

The couple of times I've had something I wanted to do that needed an SDR more powerful than than an RTL-SDR and decided I wanted a LimeSDR or LimeSDR mini...the lead time on them has been such that it would not arrive in time for the project I wanted it for.

Right now, both LimeSDR and LimeSDR mini orders place now don't ship until Feb 28 2022, except for the LimeSDR PCIe which ships Sep 29 2021. (You might be able to get the USB LimeSDR sooner from Sparkfun. They say they might have some coming in early October).

Projects where I'll need something like that are rare enough that I can't really justifying buying one just in case another such project comes up.

It is very depressing indeed. Has eBay or second hand given good results on this to anyone?

I guess we'll have to wait for https://www.crowdsupply.com/krakenrf/krakensdr but I can't imagine the price will be very light-hobby-adapted.

We are shooting for a price of $299 for KrakenSDR.

Oh thanks for your answer, that's very interesting! For someone wanting to upgrade from rtl-sdr that seems a good step up.

While you're here, I'm particularly interested whether you're aiming for something more ambitious some day or 100% focused on kraken? I'm thinking coherent 32/64 rx channels or more? Also, is there a way to assemble two krakens and synchronize them somehow so as to have 10x coherent rx? :-)

There's probably a better channel of communication for such questions ...

Thanks again for jumping in with a correction.

We have plans to offer a few different versions of KrakenSDR. The entry level version will likely be the current version with 5 tuners. We are laying out an idea for 8 tuners, integrated CPU, and pre-set antennas; basically a turnkey system for remote operations. We can definitely produce more exotic beasts, but I don't think there is much of a market for a Kraken with a hundred heads.

Ah thanks. Yeah I was more going on the side of 'build your own phased array' and (I think) needed coherent receivers for that. 299usd is an amazing price for 5 channel rx. And, since the schematics and everything (?) is open, it might be a 1-semester EE end of studies project to extend it to 32 or more :-)

Thanks for taking time to answer my questions. This project looks amazing, can't wait HW availability.

Although the direction-finding and passive radar software are both open source, the hardware is not (so no schematics, layout, BOM). Just wanted to clarify that for you and anyone else reading this thread.

It’s tough for small groups or hobbyists to do because building and keeping stock is a considerable investment and risk so a lot of times they’ll do batches constantly.

Also if you’ve had multiple projects that would want to use it why not order it knowing you’ll probably have a similar idea in the future?

Components for producing SDRs are impacted by the overall semiconductor shortage very significantly, too: keeping stock of what are fairly niche products is difficult when times are good, but with component lead times in the 52-80week ranges (and still increasing), there's absolutely no room to adapt to what customers might want a year from now. It really sucks for everyone.

Not sure your exact use-case but if you are frequency-limited by the RTL and you feel like soldering, you can mod a SUP-2400[0] to be a 4.5GHz downconverter for the RTL[1].

[0]: https://www.solidsignal.com/zinwell-directv-b-band-converter...

[1]: https://www.kd0cq.com/2016/04/part-ii-the-mod-receive-up-to-...

Does anyone know of a solution that can go down to the 5khz range? Older (...and even newer) gym equipment uses the gymlink protocol for heart rate monitors instead of low power Bluetooth and friends, and very few monitors support it anymore. I wanted to try to make a converter but couldn't find anything to output in that frequency.

I think it's pretty easy to make circuitry to shift a radio's frequency. If you did that with LimeSDR you'd still get the same bandwidth but it would be shifted into a different range.

At 5000 Hz, you could probably just rig up a sound card with a mic input.

I need to output the 5khz, not input it. Would I be able to do that with a sound card in a small form factor? The idea is to have a small box that can convert my heart rate monitor to something that the gym equipment can understand.

I believe attaching an antenna to the card's output could work. For small form factor, a USB sound card and a Raspberry Pi is a possibility.

I've used a cell phone as an audio signal generator for testing transmitters. There are apps for sending a continuous signal, and I assume it would be easy enough to write one for sending a modulated signal.

BTW, this is one reason why I won't buy a phone without an audio jack.

Ham It Plus can go upconvert as low as 300Hz, compatible with any SDR you just shift the frequency in software https://support.nooelec.com/hc/en-us/articles/360005812474

Guess I didn't make it clear and that's my bad. I need to output 5khz, not input. The input is a more modern Bluetooth protocol.

sorry maybe we misunderstand each other, I just got my license last year but this FAQ link reads, "Is the Ham It Up capable of transmitting as well as receiving? Even on lower frequencies? -- Yes, it is bidirectional, so you can transmit and receive on any frequency within range, provided that you have a compatible transceiver, like the HackRF One. "

Why not just use audio out then?

The only thing I can think of that goes that low out of the box is the old Ettus Research USRP1, B100/B110, or E100/E110 SDRs combined with both a LFRX daughtercard and LFTX daughtercard, which are DC - 30 MHz.

The SDRs themselves can occasionally be found on eBay and the daughtercards are still sold directly by Ettus Research or themselves occasionally can be occasionally found on eBay.

5KHz is pretty low, it would be difficult finding anything in the ham category because no ham bands are that low. You would need to probably build your own receiver. The good news is that 5KHz is super low frequency and could easily be received with a crystal detector. Antenna connects to an LC resonator centered at 5.3KHz, pass the result of this through a diode and then into a microcontroller. Many have analog input pins and analog comparators built-in that can trigger an interrupt. You would need to figure out what voltage level you expect to see from the detector when a heart beat signal is sent and then apply that voltage to the other side of the comparator. Some microcontrollers even have a real analog voltage output that could allow you to adjust the voltage through software. The microcontroller would then trigger an interrupt and do something every time a heart beat signal is received (send it through serial to a PC, compute heart rate, display something on a LED screen, etc).

If you need to broadcast at 5.3KHz, that's pretty easy too. There are numerous designs out there for a basic CW transmitter (https://makerf.com/posts/ten-minute-transmitter). You may have trouble finding a crystal centered at your 5.3KHz frequency.

Try an upconverter like this https://www.nooelec.com/store/ham-it-up.html or https://airspy.com/spyverter-r2/

I find totally rare that these devices works at 5kHz. instead of 5MHz or 500Khz (for example). At 5kHz you would need in theory huge antennas to being able to transmit and receive the signal correctly, and hundreds of kHz frequencies are totally achievable even with the most basics mid 1900s telecom systems.

If you know the data to send, and you are totally sure the data is sent electromagnetically at 5kHz, you can modulate in as sound in a computer and play it in a PA system connected to a long cable,

I was super into the ham world many many moons ago (DE K2KD), and my curiosity is definitely piqued with the arrival of SDRs.

Dumb question(s): Is it possible to do "traditional" HF/VHF ham stuff with these mini SDRs? Is there a good software package that emulates the old-school rigs (but with fancy new features)? Are these radios powerful enough to drive an amp to get to something like 100W (or 1500W) on HF?

Not without paying a bunch for bandpass filters and power amps to get it up high enough to actually drive your 100w amp. Most SDRs are just the transceiver without all the front-end parts you really need for a "usable" radio.

Yes, true, you'll need a bunch of intermediary stage amps as well, you won't drive 100W without something in the middle. Typically a set of usable stages would be an SDR, a very sharp bandpass filter, a 1W or so buffer, then something on the order of 5 to 10W and then into real power stages 50-100 or so and if you want to go substantially higher than that you'll need yet another stage. And each of those will need to be perfectly regulated to avoid dumping a bunch of harmonics into the air.

That's correct, though getting those is possible, and something of a fun project. I've found the stuff 60dbmco ( https://www.ebay.com/sch/60dbmco/m.html?_nkw=&_armrs=1&_ipg=... ) makes to be useful for this .

You're probably not saving any money this way, but you do learn a lot.

Before you start pumping out that kind of power I'd invest in a spectrum analyzer (which are surprisingly affordable today) to ensure that the signal that goes out is clean.

Those mini SDRs when used for transmission will output a ton of harmonics because they are almost by definition not going to have any filtering on their output because that would limit their usability. When used for reception that is not a problem, if you start using them for transmission it's a completely different matter.

Keep in mind that a power stage that responds well to a particular signal absent filtering is going to respond quite well to the main harmonics too!

That's for sure. Here's what an Ettus B210 (now $1472, not cheap) looks like with a fundamental at 145 MHz. 3rd harmonic is only 10 dB down. Power was 0 dBm (with a 20 dB pad on the analyzer input).

https://www.w6rz.net/harmonic.png

That's even worse than I thought it would be. A cheap one will likely not do better. At least they picked the HAM bands smart so harmonics tend to end up in the next HAM allocated frequency range up. But SDRs being what they are there are no passive components on board to ensure the signal stays 'in its lane'.

It's inherently a broadband device, with no output filtering. No surprise that the output is full of harmonics. I built several application specific RF front ends for Ettus SDR's at a previous job, it's mandatory for any kind of professional use.

As far as transmitting, it's not really practical. Receiving, yes. What I normally do is look for signals using my SDR, and when I want to make a contact, use my FT991A. Even though the 991 has a SDR in it, the dedicated ones have a much wider bandwidth, and the better display and UI of a computer make things a lot easier.

Check out the Hermes Lite, it's an SDR with a transmit ability, and some other additional features. I think it's 10w by default, but a lot of people put that right into an amp to get up to 100+ watts.

I immediately hit reply to suggest the Hermes Lite also; despite the crazy supply chain situation, there is a current manufacturing run going right now. With N2ADR's filter board, you have a complete 10W HF direct sampling transceiver. Add a hardrock PA and you can take that easily to 50W or 500W at an insanely low price point.

I hand build a few of the early prototypes and have a couple of the production versions too. I cannot emphasize how underrepresented I think this radio is compared to a lot of the other SDR hardware out there. It compares favorably in performance and vastly exceeds the features of most amateur radio HF transceivers in the <= $2500 range.

Because it speaks the OpenHPSDR protocol it can also run a digital predistortion algorithm called PureSignal that samples the PA output and distorts the input signal to correct for amplifier phase and power nonlinearity. Digital predistortion is relatively common in modern radios like cellular baseband, but to my knowledge PureSignal is the only open source implementation of such and the only implementation publicly available for an HF radio.

AFAIK that radios aren't usually that powerful (at least, legally to non-commercial use), and usually amplifies the modulated signal from the machine, they doesn't have an input of a direct signal to amplify and send.

Anyway, I think you can send signals in baseband from the computer or the SDR and let the radio modulate it.

One of my concerns with using an SDR is that the FCC basically requires DRM[1]:

> ...importers and marketers should ensure that [two-way radios] are properly certified and labeled as FCC-compliant and cannot be easily modified to operate outside its grant of certification.

So if you use an SDR for LoRa, you have to make it hard for users to modify the code to emit a different frequency.

[1]: https://docs.fcc.gov/public/attachments/DA-18-980A1.pdf

But this is a fully generic SDR. Can this even be certified as some consumer product?

I mean if you used an SDR in your consumer product. Let's say you make a dual purpose LoRa / WiFi radio to put in vehicles for a fleet management solution. If you use an SDR, you'd have to lock it down with DRM to prevent end users from repurposing it.

Well, it would also cost at least 10x what a dual purpose LoRA/WiFi radio would cost if it were built out of LoRa parts from Semtech and an ESP-32.

Using an SDR for this kind of product would be horribly cost and time inefficient. You would design it with off the shelf WLAN and LoRa ICs.

> If you use an SDR, you'd have to lock it down with DRM to prevent end users from repurposing it.

Not any more so than a product you designed that didn't use an SDR as its core.

Yes, but the same applies to your WiFi router even if it has hardware filters. Its simply a certification requirement.

That doesn't have anything to do with LimeSDR. This is already not a product that is certified for anything. It exists in a very gray zone where anyone using it for transmission is likely in violation of the law unless they obtained a development/testing license in the specific band they are using.

The easiest way to leave the gray zone would be to get a ham radio license. Although this might vary from country to country, generally you are allowed to use selfbuilt transmitters in specific frequency ranges, which also includes noncertified hardware.

I'm not too sure about that. Having a HAM radio license would allow you to transmit on bands licensed for HAM usage. A lot of the stuff that people want SDRs for (especially for transmitting) lies outside of those bands.

> likely in violation of the law...development/testing license

Or they just have a ham radio license

In a Faraday cage, no one can hear you transmit...

Or put the unit in a faraday cage to carry out testing. I believe a lot of people do this to allow them to play with transmission while not breaking any laws.

This is probably true, meaning you can't really have a FCC certified tx-capable SDR. However in order to legally use the tx-capabilities of an SDR you probably have to be a ham anyway. And hams don't need their equipment to be FCC certified if I am not mistaken.

> However in order to legally use the tx-capabilities of an SDR you probably have to be a ham anyway.

You can transmit in the ISM bands without a license. There are still power restrictions which most SDRs won't exceed anyways.

Yes, a general user/consumer can transmit in the ISM band without a license, but they are still required to use intentional transmitters which have been certified by the FCC.

> but they are still required to use intentional transmitters which have been certified by the FCC.

For a hobbyist experimenting with an SDR, this wouldn't apply.

That document (DA 18-980) talks about "importing two-way VHF/UHF radios... used primarily for short-distance, two-way voice communications."

In other words: cheap walkie-talkies that don't comply with FCC regulations.

Such radios typically max out at 5w.

The LimeSDR claims 10dBm - which I believe means about 10milliwatts.

In other words, that FCC memo is concerned with radios being sold that are something like five hundred times as powerful as a LimeSDR. And therefore much more likely to interfere with other transmissions.

> That document (DA 18-980) talks about "importing two-way VHF/UHF radios... used primarily for short-distance, two-way voice communications."

I thought it was more along the lines of going after Baofengs, not CB radios, hence the calling out of VHF/UHF radios specifically.

I haven't used a LimeSDR but I have used the mini version which is just about the same hardware minus the extra TX/RX channels. It's a great little device that covers a massive spectrum with a decent bandwidth as well. Comparable devices from Ettus are much more expensive. I'd love to own one for ham stuff but haven't had much time for that hobby lately. The Lime rep did give me a mini PCB on a keychain once, coolest trinket I've ever gotten at a conference.

Any suggestions on where to start to use this as a LTE uplink that reads from an SD card and uploads images? I'd like to replace my expensive faulty game camera subscription.

Using an SDR is the most painful and power-hungry way to replicate what already exists as cheap, optimized, small, commodity LTE modems.

Use the SDR for standards that don't exist as finished silicon yet.

In addition to what myself248 said:

In case you aren't familiar with the details of consumer RF standards (including IEEE 802.11, but especially LTE or anything cellular related) or don't have some prior RF knowledge: To put it mildly, LTE is not an easy standard to implement on an SDR, especially if it's your contact with SDRs. Even GSM is nowhere near a simple stack.

If you want to get a feeling for what kind of cellular projects are possible using SDRs from a hobbyist/free software perspective, have a look at the Osmocom (Open Source Mobile Communications) projects [1] as well as Harald Welte's blog [2] (one of the main Osmocom contributors). The Osmocom projects are focused on the carrier side of LTE/GSM networks AFAIK.

Another warning: While cellular technology involves lots of networks, it is completely different from the standards and terminology used in the Internet. For example, citing the description for OsmoHNBGW: "An Open Source implementation of a HNB-GW (HomeNodeB-Gateway), implementing the Iuh, IuCS and IuPS interfaces. It aggregates the Iuh links from femtocells (hNodeBs) and presents them as regular IuCS and IuPS towards MSC and SGSN (such as OsmoMSC and OsmoSGSN). It uses M3UA as signaling transport." The description totally makes sense to an insider, but if you only dabbled with TCP/IP, routers and switches so far, be prepared for learning a whole new world of networking terms.

If you are familiar with RF/cellular engineering and/or if you're trying to do this on a corporate budget then you probably already know what I mentioned above, and your project can certainly be realized for $$$.

If you're looking for a one-off DIY solution then you could use something like a u-blox SARA-R4 [3] modem (for NB-IoT/low bandwidth). Similar boards with higher bandwidth also exist.

[1] https://osmocom.org/projects

[2] https://laforge.gnumonks.org/blog/

[3] https://www.sparkfun.com/products/14997

> If you want to get a feeling for what kind of cellular projects are possible using SDRs from a hobbyist/free software perspective, have a look at the Osmocom (Open Source Mobile Communications) projects

Just want to say these guys are absolutely awesome. I got a great hobbyist discount on some hardware from them for a project I’ve admittedly fallen behind on.

My only complaint was lack of documentation, but the source code of their stuff has been reasonable enough for me to read and get stuff working.

I appreciate the insight, thank you!

https://www.drti.com/

https://web.archive.org/web/20161106083817/http://www.drti.c...

The true master of SDR....

One of many Govt focused RF companies. Start from Watkins Johnson and look at the spinoffs, although they've mostly absorbed into Bigger defense contractors at this point Boeing bought DRT in '07?

This site has a good history http://blackradios.terryo.org/index.html

I would suggest using limesdr mini with sdrangel. Lots of people working on using them for digital TV station transmition.

LTE, GSM... Does it means it is now easier to build an entirely free device that can be used as a cellphone?

You mean with proper licensing? That might be the tricky part; use of the rf spectrum is probably monitored and regulated, depending on where you're at in the world.

I've got a LimeSDR Mini. I mainly use it as an ATSC 1.0 signal generator from time to time for testing TV tuner code at work, and it does a good job, as long as I'm running it on a PC with a high-bandwidth USB 3.0 port. I've had issues going through hubs.

The multichannel components LimeSDR used are more constructed for streaming video applications, but they dont have good sensitivity in the HF range.

I wish I had gotten a hackRf instead, which also had better tooling support

This doesn't seem right, you might be thinking of a different product. The LimeSDR uses the LMS7002 IC as an all in one dual channel RF to bits chip. The chip specifically is targeting broadband wireless communications applications. Data sheet linked below.

https://github.com/myriadrf/LMS7002M-docs

The issue is not with the chip itself, but with with the choices made for the board construction https://wiki.myriadrf.org/LimeSDR_HF_Performance

I was more specifically responding to your comment that it was constructed for "streaming video applications". I don't think the SDR is targeting this application.

Does anyone know if this is more appropriate than the PlutoSDR for hobbyists? Seems like they have similar specs and are around the same price.

I haven't played with the Pluto, but I do have a USRP, which uses the Analog Devices front-end, and it's leagues more reliable than the LimeSDR.

I would absolutely go for the Pluto if it'll work for your application. I guess the main limitation is USB2 for getting the captured signal out, so you wouldn't be able to use it for crazy high bandwidth applications.

still havent used mine. supposedly theres not as much software thats ready for limesdr, that this is a mire rareified pick than many ither options.

No WiFi support?

WiFi is hard to support, it's very complex from a signal processing and timing perspective.

And probably also the protocol.

I mean you have to pay way to much money just to look at the specs (or settle for an older one that’s a pain in the ass to find for some reason)

Wifi is a pretty difficult protocol, but fortunately Linux ships with mac80211. So all that is needed is an SDR based PHY such as https://github.com/Nuand/bladeRF-wiphy/ .

Reply on news.ycombinator.com