When Analog Devices released their SDR transciever AD9361 in 2013 – it was a revolution in digital radio. SDR’s were there before, but only now you can have it all: 2 channels for TX and RX with onboard 12-bit DAC/ADCs with 56MHz of RF simultanious bandwidth, local oscillators, mixers and LNA – all working in the range from 70 (TX from 47) to 6000Mhz. Using AD9361 out of the box one could implement almost any useful digital radio, with the rare exceptions of UWB and 60GHz. You only need to add data source/sink (which is still often an FPGA), external filters and PA if your task requires it.
Finally I was able to take a look inside and peek at manufacturing cost of a microelectronic device with such an exceptional added value.
This is a little over my head, but I love the pretty pictures.
Coincidentally, I just got back today from a business trip deploying the SDR software I built to use in car races. Their existing setup was using almost a hundred individually wired hardware radios tuned to each car, but using two wideband SDR radios with my software and a high core server, they get the same data using software algorithms to handle all the radio processing.
It’s cool tech, the low level nature of it is really up my alley.
Edited 2018-05-28 00:41 UTC
Indy?
Nascar
Very interesting, that sounds like you have implemented ‘even more software SDR’ than this chip in this article, which seems to be a well designed ASIC. Are you doing signal processing on x86 hardware or something?
Anon,
Yes, highly multicore x86 hardware. In this case the RF streams came in as complex quadrature IQ samples (which had already been down converted by the hardware), from there mathematics and software algorithms take over. I learned so much about RF and I found it fascinating! It’s nice to have these diamonds in the rough every once and a while. This project was actually completed a couple of years ago, however it was recently transferred and my visit this week was to train new users.
“Software Defined Radio Radio”?
Is that something like ATM machine? 
zima,
Haha, you’ve got me scratching my head. Seriously I’m not sure if my usage was incorrect:
“Their existing setup was using almost a hundred individually wired hardware radios tuned to each car, but using two wideband SDR radios with my software…”
versus
“Their existing setup was using almost a hundred individually wired hardware radios tuned to each car, but using two wideband SDRs with my software…”
In my head SDR refers to the concept of a software defined radio interface rather than the physical radio itself, and so the second version with “SDRs” doesn’t quite read as smoothly as “SDR radios” in my mind. Even if I expand SDR as follows…
“Their existing setup was using almost a hundred individually wired hardware radios tuned to each car, but using two wideband Software-Defined-Radio radios with my software…”
… it still seems better than omitting radios. SDR radio clearly refers to a physical incarnation of the software defined radio concept. No? Oh well, luckily for me nobody is pedantic enough to dissect my posts for such things
Well, I was pedantic enough…
(if you can’t be pedantic on OSAlert, where can you be? 
)
https://www.nasa.gov/directorates/heo/scan/engineering/technology/tx…
What do you mean?
SDRs come in all shapes and sizes from tiny USB sticks to more elaborate setups with FPGA processors, etc.
SDRs are technically simpler than the RF technologies that preceded them because you no longer need complex analog circuits to process signals, which IMHO is harder to design than software algorithms. An SDR with a sufficiently fast processor enables us to get rid of a lot of analog circuitry. One analog frequency shifter and A2D converter is all you absolutely need to have a viable SDR solution.
SDRs are supper cheap in commercial quantities if you are willing/able to hack off the shelf products:
https://www.rtl-sdr.com/tzumi-magictv-wifi-tv-tuner-device-contains-…
Here’s some more SDR info.
https://spectrum.ieee.org/geek-life/hands-on/a-40-softwaredefined-ra…
Those who want an actual developer kit are generally looking at hundreds of dollars, which may also get you better specs, but there’s no need to spend a lot of money to start playing around with SDR.
Edited 2018-05-28 14:10 UTC
I mean getting into the SDR market for hardened environments like the ISS. It sounds like a slightly easier and cheaper route to get a NASA supply contract with these things than having to own/start a rocket company.
May you recommend a good and cheap antenna? I have friends that have farms and, perhaps, this would allow them to have a cheap way to connect their facilities. Right now most of them have one point for satellite reception of Internet/Phone/TV and it would be cool to distribute the signals around their properties.
acrobar,
It sounds like they could use standard WISP (wireless ISP) gear that support standard WiFi protocols. Ubiquity is the most popular, although there are many vendors with cheaper equipment (yet still compatible).
https://www.ubnt.com/products/
https://www.tp-link.com/us/business-networking/wireless/outdoor-acce…
Would they need point to point links? Yagi and dish antennas will have the longest range but have to be pointed directly at one another. I’d probably look at one or more sector antennas to cover a large field (like those in cell phone towers). They will work better than standard unidirectional antennas that most residential WiFi routers come with.
I would think a 13dBi sector antenna is a good compromise between range and coverage for a field. With an ideal setup, you could expect several kilometers, however many people don’t factor in what’s called a fresnel zone. A significant portion of the radio energy travels around your line of sight, a picture will make this clearer…
http://afar.net/fresnel-zone-calculator/
https://www.everythingrf.com/rf-calculators/fresnel-zone-calculator
So at 2.4GHz and 1 mile distance, the path would ideally be 24 feet higher than ground level. You can reduce the height and it will work, but the loss of signal strength reduces the range.
While standard Wifi operates at 2.4/5GHz, lower frequencies are typically better at getting around objects like trees and buildings. Since antennas are tuned for the frequencies in use, make sure you get an antenna that matches the frequency(ies) you intend to use. Antennas become significantly less efficient at the wrong frequencies.
I’m curious how things will work out!
Edited 2018-05-28 17:48 UTC
Hey, thank you! I will take a look on links you provided. I will have to check with them but, from memory, the worst case was something like 2 to 3 miles from the “headquarter” (actually, their homes). Generally, things are not aligned and they were looking for something omni. For them, the most important things were phone services, security and data transfer (about whatever they produce there). Of course, the equipments must resist the rough conditions where they may be installed on. For sure, they are not rich and, as so, cost is a very important aspect of it.
acobar,
Yeah, consumer routers aren’t great for outdoors, but WISP equipment is designed to be weather proof and can likely withstand higher levels of static discharge too.
I’m not rich either, which is why I personally buy off-brand and/or used equipment. If I had a large budget, I’d be able to build bigger things, haha.
An omnidirectional antenna will perform worse than a few directional sector antennas. If you look at cell phone towers, they’re comprised of an array of sector antennas, each pointing in one general direction. Even with a low budget, I’d still recommend going this route over omnidirectional for a better signal.
https://technogog.com/information/how-stuff-works-cell-phone-towers/
Edited 2018-05-28 20:10 UTC
This guy is probably living in fantasy land where companies do things for free or for the cost of raw materials and have no other expenses whatsoever – companies are created to be profitable, not to sell things cheaply for specific person. So what there are some R&D+”other” costs for existing chip ? Manufacturer and distributors need to make money to invest in the future designs, trainings, marketing, salaries, etc. Investors need payback for their investment. Even failed research projects need to be paid for. If they don’t sell anything or sales are not great, they also need money just to keep to company financially stable.
Thank you https://www.goztepetemizlik.com.tr/ofis-buro-temizligi/ for informative article