POLL: What do you want in a future low cost SDR?
We've decided to ask the community what sort of radio would be more popular for a ~$30 RX-only SDR. Note that thus far this is only a hypothetical SDR that does not yet have any designs. We are just feeling for what's most interesting to people and exploring for future ideas which may not even be feasible at this time.
As always with engineering, there is always a trade off. It is likely that any low cost SDR can only be wide band, with basic RF performance, narrowband with good RF performance, or expensive.
Loading ...Wider bandwidth: Currently the RTL-SDR allows you to see up to 3.2 MHz of spectrum live. Higher bandwidth (10 MHz or more) would mean seeing more of the spectrum at once without needing to retune, faster whole spectrum scans and the ability to receive wideband signals.
RF Performance: The RTL-SDR has an 8-bit ADC. While sufficient for many applications, in the presence of strong signals the ADC will saturate resulting in poor reception and signal overload. Higher end SDRs like the SDRplay, Airspy, LimeSDR etc use 12-bit or higher ADCs.
Frequency Range: The RTL-SDR V3 can tune from 500 kHz to 1.75 GHz. An expanded SDR could potentially tune up to 4 GHz or even higher.
If you're interested in other things, please comment on this post!
I want a SDR that can be prototyped with a USB device and a Raspberry Pi – then licensed for production in a commercial device.
After using SDR such as the CIAO H101 and the RTLSDR I can say for something cheap a tuneable preselector which is cheap enough to provide, which means a larger hardware case, will help to rid us of out of band images from Mediumwave and other bands, especially above 14 MHz such as the HF upconverted RTLSDR suffers from. This would aid in both antenna tuning or peaking and out of band rejection. If you add a larger case you can also ad in some chokes on the USB input to decrease some PC noise. This would be excellent for making the RTLSDR devices perform better.
To have all 3 sounds tempting ,but I assume unfortunately to expensive or you wouldn’t have asked where to put the preference on .Having a hardware that is supported in the available programms is also a good idea.
Other ideas that come to mind for an improved functionality,
– add a switchable (Pin-)Diode attenuator to allow to detect/reduce IM
– ext. synchronisation add cut/soldering point for
– pcb design to allow optional metall shielding
The Blog has always been e reference point for novices, for informations and for products supply, however you are neglecting this market share lately. We haven’t seen new products released for the novice. How many times we have been talking about a HF wide band loop antenna amplifier, just to say one?
Developing higher level projects like the KerberosSDR doesn’t have to stop the release of simple circuits for the beginner or you will loose the supremacy in your traditional market share; this means opening a gap with less popularity for the Blog, less visits and people being loyal to nothing but the chinese gadget vendors.
All of these items listed are not mutually exclusive. I would like to see low-cost sdr’s address all of thee items.
I would like to see a 12 bit+ version with good HF band filtering. Good software and OS compatibility essential. No need for VHF and up coverage. Price it 3- 50 $ and it will sell like hot-buns
Ask these sort of questions and everyone is going to ask for Perseus levels of performance for $10 🙂
I’ve used the V3 RTL dongles a lot and they provide a lot of basic functionality at a good price. I consider them to be a good introduction to RF and a bridge or software developers to get interested in all things wireless. In fact I’ve given away RTL dongles to various friends and colleagues who have shown a slight interest in RF and think they are a bit like a modern day ‘crystal set’ for the internet generation.
Marty has hit the nail on the head when he said whatever it is, it needs to work with software. Many of the features being requested are already available in the SDR Play RSP1A, at a price point that is not that much greater than what is being proposed. However because it is a competitor to Airspy, they have chosen not to support the RSP in SDR Sharp.
The great advantage of the V3 is that it is cheap, small enough to carry around for casual listening with a laptop whilst travelling (but don’t get arrested in countries that are problematic), and sufficiently versatile that it can be used for various projects that would otherwise be prohibitively expensive e.g. Radio Astronomy, Direction Finding, Passive Radar and various other projects requiring the capture of multiple signals for post processing.
As another example I’ve used RTL dongles for various projects such as the Farnham WEB SDR which uses many of them to provide coverage of various amateur bands, ranging from DC (almost) to 10GHz. http://farnham-sdr.com/
Based on these experiences I’d say that for me, I’d like more of the same……
The basic form and functionality of the V3 (the SDR Play contains a lot of empty space)
Sampling at >60MHz to provide DC to 30MHz coverage using direct sampling without aliasing
>5MHz of bandwidth so that a ‘decent’ chunk of spectrum can be seen at once
12bit (or more) resolution for improved dynamic range, however lots of users can’t take full advantage of this as they are using poor antennas and 8bits is often adequate.
I would not bother with extra filtering because the SDR play already provides this at a slightly higher price point, and I wouldn’t bother with frequency coverage above 2GHz. Most folks only go up to 1090MHz for ADS-B reception, and if you are interested in anything higher in frequency on the microwave bands, you would probably either use an external down-convertor with additional low noise pre-amplifiers and filtering, or go for something more suited to bi-directional communication, such as a Lime or HackRF.
Some other thoughts………..
Ethernet rather than USB would be interesting as this would allow it to be operated from anywhere with a network connection, or operated at the mast head or remotely without having to use USB extenders or similar.
The ability to GPS time stamp the digitised output stream (like the KiWi SDR) would also open up a whole range of new possibilities for distributed receive systems that could be combined and manipulated in post processing in order to provide diversity reception, noise cancellation, direction finding etc. Imagine something like a Spyserver network, where you could select multiple receivers and combine and manipulate the sources, to create something equivalent to a radio astronomy large aperture array.
Maybe a transmit only companion to the V3 with an output level of around 1mW ?
Just my thinking.
That’s allot to expect out of something proposed to be low cost, especially the GPS time stamp being thrown in there.
Must be LSM capable with above mentioned
also.
All 3 are nice features.
But the one that is essential is for the receiver to do a good job as a receiver.
I would say in order of importance:
(1) Good RF reception (high dynamic range, image rejection, etc.)
(2) Wide bandwidth (10 MHz).
(3) Wide frequency range (note: frequency range can always be expanded with up/downconverters and most signals of interest to the hobbyist fall within the range of roughly 1 MHz to 1.7 GHz, roughly the coverage of the existing dongle).
Hi,
12bit ADC is what I voted but I also want to see coverage. 1MHz to 4GHz.
Similarly if a high speed DAC can be integrated and tx/rx select then it will be a nice little sdr trx. Well this is dream I know???
Beyond my vote for better dynamic range would be a much more simplistic set of band filters that can be toggled by software. The LimeRFE is nice, but still too much product. https://limemicro.com/news/lime-microsystems-unveils-limerfe-software-definable-front-end-add-on-for-limesdr/
Would be nice to have: high pass for BCB or FM BCB notch, plus maybe a few VHF/UHF ham band filters and perhaps something for L band (somewhere). Cheap SDR for HF is not of interest to me at least, but a simple HPF could be fun. Again, make it software definable like the nice built in DC bias
Frequency Range would need Wider bandwidth, most of the signals at higher frequencies are wider
But even if it was narrow-band, fast scan and only covered the ISM bands that would be useful to anyone who uses the ISM bands.
Wider bandwidth would need better RF Performance (higher dynamic range), or external filters and controllable gains and attenuators.
RF Performance (higher dynamic range) does not really require anything else to be a good addition.
What does surprise me is that no one has yet programmed a MCU, with a builtin ADC, to emulate all the USB commands expected from a RTL2832 chip. So that the device could be 100% RTL-SDR compatible with existing software but offer larger bandwidth or even offer special new debug modes that bitpack even more bandwidth at a low lower dynamic range, or higher dynamic range at a lower bandwidth, (Like the SDRplay does 14bit at samples rates from 2MSPS to 6.048MSPS ; 12bit 6.048MSPS to 8.064MSPS ; 10bit 8.064MSPS to 9.216MSPS ; 8bit >9.216MSPS), but that is probably not even possible at the selected price point with a cheap MCU because of RAM and MIPS constraints. It would require a grey haired programming guru to even contemplate attempting something so complex. The nice thing about a layer of emulation is that the product would be still be a RTL-SDR device but with extra features available.
How about something that can receive just the ham radio bands, but really well without a ton of extra junk dangling off of the dongle? And a good bit of software designed around that function?
You’re missing probably the most important thing about SDRs…The software application that makes it operate…Either it has an app that works REALLY WELL with that SDR (like ELAD or SDRPlay) or better compatibility to the apps that are being kept up by the community…
Mast mounting capability and PoE. Airtight housing and good temperature range.
All I wish for the next low-cost SDR is enough range to receive L-band sats. There are many interesting sats around and slightly aboce 1.7GHz, but tuning to their frequency starts overheating the SDR to the point where it’s risking damage.
Image und spurri suppresion at highest level… no more USB Noise at 480/240/120MHz….
Low frequencies without switching modes.
Better selectable front end filtering.
The ceiling price sounds more realistic up to 60-80 USD. Although the value should be among the questions.
Multiple antennas for array processing capability. I would like to have an antenna array with at least 4 elements whouse beam can be steered by software. This would allow to receive LEO satellites without a rotator. Kerberos is a beggining but this is a path that must grow in the future. Thanks.
have the three options by spending a little more
Higher frequency range only makes sense when paired with higher sample rate/bandwidth— most interesting signals above 2ghz are wide band (e.g. WiFi, LTE, GSM, CDMA) or spread spectrum (Bluetooth).
I wish we could have all these options together, for 30usd…
🙂
Sure and for 50 USD bundled with a Porsche as “high speed version” ?
Put me down for 2 of the High Speed Versions.