REGISTER | LOGIN
Breaking News
News & Analysis

5G for Command & Control

Running on Phones Just Sideline
10/6/2015 10:00 AM EDT
6 comments
NO RATINGS
More Related Links
View Comments: Newest First | Oldest First | Threaded View
R_Colin_Johnson
User Rank
Author
Re: latency
R_Colin_Johnson   10/7/2015 4:12:33 PM
NO RATINGS
His plan is to put smarts into the base stations so that the are preloaded as you move around so that the information you need is always available nearby, Dee. Don't shoot the messanger--that is his plan, and I'm putting it together into a slideshow for all to scrutinize. Thanks for the comment, always a pleasure to hear from you.

DeeCee430
User Rank
Author
latency
DeeCee430   10/7/2015 12:42:02 PM
NO RATINGS
"One millisecond latency no matter how far the signal travels" is of course impossible, unless he has discovered how to send signals much faster than the speed of light.  Light in a vacuum only travels 186 miles in 1 msec., and even less in a copper wire or fiber optic cable.  And lots of other things add to latency.

R_Colin_Johnson
User Rank
Author
Re: The premise is wrong
R_Colin_Johnson   10/7/2015 1:51:23 AM
NO RATINGS
realjjj: I like your being true to your handle with "We do need to operate in the real world, unfortunately." Yes, if 5G were just about speed, it could be here much faster, but the reaity is that low latency and making basestations smarter is a integral part. And since the safety concerns of driverless automobiles and such are a long way off, there seems to be no hurry to develop it. After all, 4G is still gaining penetration! That's why I as so surprized that TU Dresden was able to talk the NFL into doing a live test of their 5G implementation this year!

realjjj
User Rank
CEO
Re: The premise is wrong
realjjj   10/6/2015 2:41:25 PM
NO RATINGS
We do need to operate in the real world, unfortunately.

There is this huge difference in volumes between 4G/5G connected IoT and all IoT, for the next 10 years (without radical changes) the majority of 4G/5G will be smartphones and post-smartphone (new form factors that replace it).  So i am not so much disputing total IoT projections just the claim that 5G has much to do with those kind of numbers and IoT.

5G can't be a niche for just cars and some other medium or low volume devices. Yearly car sales are maybe some 80 million , the US is maybe some 17 million,, China 23 or so - ofc one also must factor in the attach rate for connectivity in cars. Cars is a broad subject and i would rather not go into that but our selfdriving cars should be fully autonomous and not depend on any external connectivity to be safe and function properly. Connectivity would only enhance it's awareness but without being fundamental. So ,as i see it, there would be just a small window for connectivity to be crucial in cars, before cars become good enough. After that the relevance of connectivity - in navigation- would decline.

Sure low latency is a plus in just about anything and maybe they have some interesting solutions. Encryption and compression (maybe even including memory compression here) are increasingly important so good ideas wouldn't hurt.

R_Colin_Johnson
User Rank
Author
Re: The premise is wrong
R_Colin_Johnson   10/6/2015 1:01:06 PM
NO RATINGS
realjjj:You have a real talent for bringing the analysts back down to earth, thanks for keeping it real. Those numbers come from  people who make their living off predictions, so are inclined to be inflated--Gartner for instance predicts 26 billion by 2020, ABI says 40 billion by 2020, Cisco says 50 billion by 2020 "on its way to 500 billion" which IHS confirms predicting 75 billion by 2025. But the only ones or those that can't be served by wired connections are mobile devices, and most of them will be fine with 4G. The reason for 5G is that some of them will need the one millisecond guaranteed latency and will be willing to pay fo it. The NFL will be the first this winter to test prototypes that sync all the monitors in a stadium, but it will really be car safety and the like that will be willing to pay for the safety element provided by 5G. Thanks again for keeping it real. I always enjoy your comments.

realjjj
User Rank
CEO
The premise is wrong
realjjj   10/6/2015 11:02:53 AM
NO RATINGS
50 billion to 500 billion by 2020?

Smartphones, tabs, watches , PCs, TVs ,speakers ,glasses and lightbulbs are the bulk of it in consumer. Volumes for these are somewhat easy to measure.. Outside smartphones a very very small % will have 5G due to costs. In the 2020-2025 timeframe very few products using 5G can gain global traction in consumer due to costs, as long as carriers are allowed to exist anyway. Sure govs and businesses are more likely to waste huge amounts of money by using 5G but without consumer, volumes remain limited.

So unless govs get a clue and radically change how spectrum is awarded and how networks are built and operated, 5G has minimal relevance in IoT and the only goal should be to cut costs. Even in developed markets some 80% of data used by smartphone users is over wifi, 4G is just a very costly redundancy. Even smartphone adoption has hit the data cost wall. Look at how global internet access growth is slowing down, that's all on data costs,not much else.

In a sane world we might need low latency for 50 billion devices, in this one not so much.

 

Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed