Rick, how is 5G defined? 4G is 1 Gb/s throughput for stationary links I think, what is the corresponding number for 5G? and why do we need 5G? isn't 1 Gb/s enough? do we need to watch HD movies on msart phones? Kris
Interesting question. Very few, if any, could have foreseen where we would be today, let's say 50 years ago in 1964. But I think we might be better equipped to approach the question going forward.
How much data do we need local access to? For instance, would it even be desirable to have all financial transactions in the world, on a real time basis, go through our personal mobile device? It would need to be a very powerful mobile device to make use of such enormous quantities of data. I'm sure we could find more efficient ways to get the desired results.
The real question is how much unique information could be available to a mobile device that would be useful to transmit? These could include biomentric information, precise location and orientation information and various user inputs like voice, camera, keyboard etc. And what information needs to be received to present, control, or otherwise be useful.
Another is that each mobile device, or other categories, could be a peer to peer part of some sort of mesh type network. But even here, maybe especially, each node would carry a small part of aggrigate traffic.
I believe the future will bring better understanding of how much data personal mobile devices need and there will be a useful limit, not defined by RF capability but by utility.
The way I see there are two hard limits eventually. First is the amount of bandwidth human is capable of absorbing (we only have so many eyes, ears, and bran cells). You can see young people working on laptops with multiple windows, listening to the music and texting on their phones at the same time. They look to me close to their bit absorbing capability already. Second is the physics of wireless communication. The mount of spectrum is fixed and just can't continue going up with frequency as teh signal attenuation beyond 60 GHz is huge. Yes, you can do beam forming, play coding tricks, design better modulation schemes, etc. but there is probably a limit to that too.
I don't think we should underestimate beam forming as a potential approach to alleviating spectrum limitations. The technology can get rather esoteric but we really haven't begun to explore synthetic aperatures at the base station level, nevermind with the mobile devices themselves.
Beam forming is an interesting technology and has its place for use for sure...but let's keep in mind it has been known and used for years in military applications...plus you can send a beam in one direction...I don't see how this can really improve throughput in 360 degree circle...Kris
I was was going beyond simple beam formation and raising using a synthetic aperature. Not only can a 'beam' be steered but also focussed. The term 'beam' is not be adequate for such spot delivery. It could be several beams, from several physical locations, where only at the target destination would the waves be in proper phase. To accomplish this would require all transmitters to have absolute time accurate in the order of femto seconds.
Also, the part about forming a beam in only one direction is not true. With high accuracy DACs driving each antenna element one can digitally assign multiple beams at multiple or same frequencies. The same is true of receiving. I believe the digital signal processing would relatively simple. The DACs and ADC would need lots of bits and it would help if all amplifiers were sufficiently linear. The steering and focussing would help the signal to noise ratio.
Sure Les, that might be theorethically possible...I think I run some calculations about this with my grad students at some point in the past...the power dissipation will be enormous though...which bring me to the third limitations of the bandwidth: how much power do we really want to burn by sending those wireless signals? Kris
Of course there're implementation issues but I believe such techniques would actually reduce the required power being transmitted per channel. All the necessary phase shifting would be done digitally before being presented to the DACs associated with the particular antenna elements. The resulting beam being directed only at the intended receiver would require less transmitted power for a given received signal strength, sort of the equivalent of antenna gain. And as I said previously such techniques at base station receiver would result in improved signal to noise ratio.
The bigger picture is this is being proposed as a possible means to alleviate spectrum resource limitations. This would fall under space division multiplexing. And of course the need to find a way around spectrum limitations implies greater amounts of communications with higher throughputs, hence 5G, 6G, etc. THIS is what will tend to contribute to higher amounts of power surrounding us. But techniques such as being proposed here will also have the potential to minimize the power required.
PS, these techniques are already, in some form, being applied in modern wifi routers such as those complying with 802.11n, ac and presumably, ad.
@Kris: I don't know max LTE rates now or in future--does anyone else?
5G is still being defined. You know the drill, we always need faster, faster. I suspect folks will try to get a couple Gigs with mmwave, though increasingly capacoty is more importnat than throughput IMHO
thank you Rick...I think max LTE rate is 100 Mb/s when moving and 1 Gb/s when stationary...but this is the theory or lab at the best
we always say we need more bandwidth, but do we really need it? what for?...we don't eat more (well, some do and get obese), we don't drink more (well some do and need attend AA meetings later), we don't drive faster (the cars drive as fast as 50 years ago) and we probably have no more sex than i the past (I am guessing here ;-)...do we truly need to have more wireless bandwidth? Kris
Oh we do, krisi. Just imagine if we could move most of the computing power towards the cloud with very thin cheap and standard clients. Higher bandwidth would allow for that, and you can imagine that the sky is the limit in this case.
28 or 39 GHz will not work in all scenerios but will in many. Beam forming will be key at both the mobile end and at the base station. Also terrain mapping for reflective surfaces in urban environments as well as attenuation factors through walls.
The mobile device will need higher accuracy location mechanisms and orientation info to report to base stations so beam forming will be colabrative effort.
Seemless handoff of course, not only geography, including obstructions but bands too.
I think present GPS would be sufficient to get significant gain from realatively simple beam forming. In general though GPS is not sufficient for other uses and will at least have to be supplemented. Auto navigation needs much better accuracy, down to location within a lane for instance.
What works at the mobile level is not what's important. There will need to be much added to infrastructure to support much that is evolving. Certainly location precision is part of that. Mobile will able to meet this as the need evolves.
If and when synthetic aperatures start being used then target position accuracy would become much more important.
GPS is not adequate for indoor positioning - cellular, Wifi etc. can help with that. There is a lot of research on fusion of GPS, Cellular, Wifi etc. data for better positioning (and related services) indoors and outdoors. Lots of money to be made for the winners there...
My response to Rick was the short answer. It pertained to the relatively near future, and only when such beam forming might be necessary, and available. As far as I know there is presently no beam forming deployed that would have to know target position to any great accuracy.
As you mention, there will be adjuncts to GPS going forward, one of them being WiFi. But if WiFi were available it would make more sense to use it for data, including VoIP. Of course, handovers and tarrifing would have to be worked out.
This is an exciting time for determining what kind of research will be conducted into the future. High-frequency roaming is one important challenge, as is linking millimeter and traditional services and dealing with skyscrapers and other blockages. But the fundamentals of the technology are already in place.
A Book For All Reasons Bernard Cole3 comments Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...