Alcatel-Lucent has developed a magical radio that can do 2G, 3G and 4G in a two-inch cube, remove unsightly antennas, reduce network costs enormously, and save the planet too. Claims like that should be greeted with scepticism, but it seems that Alcatel Lucent has achieved something quite remarkable, even if their application …
"Two Inch Cube"? Double it!
I've no problem with rounding numbers for brevity's sake. But be careful: small differences in one dimension rapidly become huge differences when cubed:
A real two in cube would have a volume of 131.096 cubic centimetres:
this 6.5 cm cube has a volume of 274.625 cc: more than double the first.
It might also double the share price from a bit over two euros.
My understanding of software-defined radio has it,
though I'm not an ee, that it is easy to try and do too much in software. Like the gn00 guys insisting that a board with a DSP and some ADCs and DACs is enough, but the guy trying to use that board for a radar finding that he needs filters --mechanical ones in his case, oh irony-- to make the thing stop overload itself with noise.
So what I would've done here is to take all the components needed in 2-through-4G systems, combine each common subsystem into one parametrised version. Exactly like you'd modularise software, only done in hardware. So it has to have at least a tunable rx/tx plus filters, and some way to, reading the article, decompress the incoming waveforms and send them out into the ether.
This is fairly smart use of the loads of bandwidth available in optical fibre, but likely does require dedicated fibre runs; you don't really want some congested router or switch to sit on the waveforms for even a couple miliseconds.
That leaves the question just how efficient all this will prove to be, plus whether future systems won't outstep the limits of this design. If 5G or 6G should move up to 60GHz, for example, I doubt this thing can keep up. But since multiple can apparently work together pretty well, gradual replacing with upgraded boxes is probably a viable option.
Oh, and apparently the USoA military already have software defined radios of sorts deployed, though details are likely classified and anyway I haven't kept up. Something for Page to comment on?
Have to say that while this looks like an interesting application there isn't anything especially new or radical involved.
In the case of the USoA things you're referring to I think you mean the Thales kit, which was originally Racal stuff developed in the UK. Not sure where the manufacturing is done now though.
Forgive me if I am wrong...
... but, what they are describing is essentially a 'micro' RBS [Radio Base Station].
All the rest of the architecture described is essentially, "normal".
The 'central server farm' is currently the NOC, [Network Ops' Centre] which is often, but not always co-located with important elements such as the HLR, MSC and of course the Charging System, either for pre-paid or post-paid subscribers.
Getting the "band width" out to RBS has always been done by Fibre, or, as some would have it, fiber. Microwave is often also used on long runs, say, the north African coast.
In Blighty, BT or similar wholesalers carry the traffic.
Next in the food chain is often the BSC, or Base Station Controller. This is often the level at which RBS to RBS handovers are done, BSC's themselves being controlled by an MSC. Pooling can save resources here, allowing underused BSC's to take traffic when fellow nodes are over worked : busy hour periods are typical.
What Alcatel have done seems to be a device which would be ideal for in building coverage, especially given the 'directional' capability.
The rest of the network seems to be largely ignored.
short answer. You're wrong.
Look up what a software defined radio is. Pay particular attention to "phased array antennas" and "synthetic beam forming".
This is *low* level, high bandwidth activity.
Arthur C.Clarke Was Right
"Alcatel-Lucent has developed a magical radio that can do 2G, 3G and 4G ...."
"Any sufficiently advanced technology is indistinguishable from magic." - Arthur C. Clarke.
"it is the lack of intelligence that's supposed to save billions of dollars, and the planet"
If that actually worked, the US Congress would have already done it. They are vastly overqualified if that's the basis for success.
People might like to look up the architecture of the Tracking & Data Relay Satellite System
These were the 2 sats deployed by the shuttle to allow (in principal) global relay of launcher and satellite telemetry back to a US ground station from *anywhere* on track. A neat idea to cut ground stations in exotic (and expensive) locations.
To get flexibility they went with a phased array of antennas but the actual *calculations* of what phase delay each element should be using (if not the *actual* delay itself) was calculated on the ground (I think their orbit is around 1100Km)
Until I'd heard about the architecture I had not thought you could do the updates fast enough. Obviously I was wrong.
Incidentally a *very* useful trick in this is to read the Nyquist sampling theorem carefully. This permits undersampling. This seems to be the key to doing SDR on Linux boxes.
- Boffins attempt to prove the UNIVERSE IS JUST A HOLOGRAM
- China building SUPERSONIC SUBMARINE that travels in a BUBBLE
- Review Raspberry Pi B+: PHWOAR, get a load of those pins
- That 8TB Seagate MONSTER? It's HERE... (You'll have to squint, 'cos there are no specs)
- Review Reg man looks through a Glass, darkly: Google's toy ploy or killer tech specs?