×

Announcing: Slashdot Deals - Explore geek apps, games, gadgets and more. (what is this?)

Thank you!

We are sorry to see you leave - Beta is different and we value the time you took to try it out. Before you decide to go, please take a look at some value-adds for Beta and learn more about it. Thank you for reading Slashdot, and for making the site better!

Huawei, Vodafone Test Out Hybrid System That Combines LTE and GSM

timothy posted about 6 months ago | from the they're-into-bonding dept.

Cellphones 46

alphadogg (971356) writes "In the hunt for more spectrum to speed up mobile networks, Vodafone and Huawei Technologies have successfully tested a technology that lets LTE and GSM share the same frequencies. The speed of future mobile networks will depend on the amount of spectrum mobile operators can get their hands on. The more they get, the wider the roads they can build. One thing they can do to get more space is to reuse frequencies that are currently used for older technologies such as GSM and 3G. But that isn't as easy as sounds, as operators still have a lot of voice and messaging traffic in those older networks. However, using a technology called GL DSS (GSM-LTE Dynamic Spectrum Sharing) Vodafone and Huawei have shown a way to allow GSM and LTE to coexist."

Sorry! There are no comments related to the filter you selected.

interesting (2)

davethomask (3685523) | about 6 months ago | (#47254657)

does this make my eee pc faster?

Re:interesting (1)

tepples (727027) | about 6 months ago | (#47254771)

Network-bound tasks
If your Atom laptop accesses the Internet through a USB adapter that connects to a UMTS network, and your carrier deploys LTE alongside UMTS using GL DSS, then a new USB adapter supporting GL DSS will make your laptop download things faster.
CPU-bound tasks
Even if you're downloading things faster, you might not notice much of this speed because it won't help your laptop process what it downloads faster, especially now that web pages are using more JavaScript. If your tasks are CPU-bound, you might have to replace your Eee PC with a Transformer Book T100 [asus.com] , the successor to the Eee PC with a quad-core Atom CPU, for more speed.

Re:interesting (1)

Anonymous Coward | about 6 months ago | (#47256207)

this is not an FBI acrinums thread!!!

ELI5 these acronyms (2)

tepples (727027) | about 6 months ago | (#47257115)

In case anybody reading this is five:

Eee PC is a brand of low-end compact laptop computer formerly manufactured by ASUS.

USB (Universal Serial Bus) is the little rectangular port in the side of your laptop, marked with a drawing of a stick figure carrying a bowling ball. (See #36 in this Photoplasty [cracked.com] .) It's where you plug in a mouse, a phone, a memory card, or an adapter the size of a memory card that connects your computer to the cellular network.

UMTS is the language that 3G (third-generation) cellular devices, such as phones and tablets, speak to connect to the Internet, except on Verizon or Sprint.

A CPU (central processing unit) is the part of a computer that processes. Processing means doing the math and making choices, such as laying out where the words and pictures go on a web page. A task is CPU-bound if the CPU can't process it fast enough to keep up with all the information coming in.

GL DSS, defined in the summary, lets 3G and 4G (fourth-generation) cellular devices share one frequency. I'm not sure how this works, but it may be done using TDMA, a fancy term for "taking turns".

Re:ELI5 these acronyms (0)

Anonymous Coward | about 6 months ago | (#47258031)

UMTS is not what you think it is. UMTS can be used for the CS domain, no internet involved.

Re:ELI5 these acronyms (2)

tepples (727027) | about 6 months ago | (#47258259)

I meant that UMTS is used for packet-switched (PS) Internet data, not that it's used for only packet-switched Internet data. Besides, I'm under the impression that a lot of devices will prefer GSM over UMTS for circuit-switched (CS) voice calls to save battery; hence the lack of 3G on the original iPhone and the "3G" switch in the Settings of iPhone 3G. Finally, to maintain the post's conceit of atonement for initialism overload, I had to oversimplify a little because of the constraint of not introducing any new initialisms in the post without defining them.

Re:interesting (0)

Anonymous Coward | about 6 months ago | (#47254873)

I think the bottleneck in said hardware is more along the lines of the CPU, limited RAM and integrated graphics rather than the internet connection. Insisting on the bleeding edge NT kernel is not going to help you a jot. In fact, you might be better off using a kernel with a smaller footprint, such as a UNIX clone.

Jussayin'.

Re:interesting (1)

bluefoxlucid (723572) | about 6 months ago | (#47254953)

Yes, and then Slashdot can send you to http://lp.getfree-soft.net/mpc... [getfree-soft.net] every 3 seconds instead of 5 seconds.

Everything should be done over WiFi (0, Interesting)

Anonymous Coward | about 6 months ago | (#47254747)

We're just wasting time screwing around with LTE, GSM, etc. VoIP over WiFi is the way to go.

Internet access for vehicle passengers (1)

tepples (727027) | about 6 months ago | (#47254795)

As I understand it, cellular data is good for 1. transit passengers, and 2. customers in shops that have chosen not to offer free Wi-Fi to customers in order to discourage loitering. How do you recommend to deploy Wi-Fi in vehicles without using GSM, UMTS, LTE, or another cellular technology?

Re:Internet access for vehicle passengers (1)

petermgreen (876956) | about 6 months ago | (#47255003)

As I understand it, cellular data is good for 1. transit passengers, and 2. customers in shops that have chosen not to offer free Wi-Fi to customers in order to discourage loitering.

3: people working on client sites where the client is too paranoid to let outsiders on their network.
4: people staying in places which either don't bother to provide wifi, provide a terrible wifi service or charge through the nose for wifi.
5: people trying to find their way arround on foot in a new city (google maps is pretty good for this, there are probablly offline alternatives but they are nowhere near as ubiquitous)

Re:Internet access for vehicle passengers (1)

Krojack (575051) | about 6 months ago | (#47256291)

I just got back from a vacation. All the hotels that offered free Wifi had it throttled from 0.5 to 1mbit. They offered faster speeds for $5/day. I just flipped on my 4G hotspot and let my family use that. I was pulling 15 up to 60mbit. Still on unlimited data via Verizon thankfully.

Also from my understanding, if there is no wifi password then the data between your devices and the wifi isn't encrypted. Correct me if I'm wrong. This is why I won't use a free wifi without passwords.

Re:Internet access for vehicle passengers (1)

Phreakiture (547094) | about 6 months ago | (#47256877)

You're sorta wrong.

The password you use to get onto a WiFi network will keep the rabble out of that network but anyone who is let onto that network will be able to read any packet on that network, because the password is used as a symmetric crypto key.

In short, if you want to avoid using public wifi because it doesn't use a password, you're avoiding it for the wrong reason. That said, using a VPN or TOR can mitigate most of that risk.

Re:Internet access for vehicle passengers (1)

petermgreen (876956) | about 6 months ago | (#47258323)

Also from my understanding, if there is no wifi password then the data between your devices and the wifi isn't encrypted. Correct me if I'm wrong. This is why I won't use a free wifi without passwords.

Most public wifi services do not have useful encryption whether they have a password or not. Lets look at the wifi encryption/authentication options as they relate to public wifi.

open wifi: no encryption
open wifi with a web based login required to unloc internet access: no encryption
wep: encyrpted but everyone with the network password has the key and trivial to crack even if you don't have the network password
wpa/wpa2 in PSK mode: encrypted but everyone with the network password has the key, with public deployments the network password is likely easilly available to an attacker and also likely short/simple enough to easilly bruteforce.
wpa/wpa2 in enterprise mode: AIUI these theoretically give inter-user protection but I wouldn't like to place bets on how secure it is in practical deployments. Also has practical difficulties that make it tricky to use for public wifi.

The bottom line is that if you want security on public wifi you should route all your traffic over a VPN with strong authentication and encryption.

Re:Internet access for vehicle passengers (1)

mythosaz (572040) | about 6 months ago | (#47257431)

RE: 5. Google Maps allows pre-caching of maps, for whatever that's worth.

Re:Internet access for vehicle passengers (1)

Wing_Zero (692394) | about 6 months ago | (#47259249)

in my experience, the it's pre-caching a ROUTE. I used it to get to the museum in Milwaukee, and we took a turn early because of construction. google maps got confused, and tried to connect to the server to update the route. Since the tablet had no cellular capabilities, it locked up maps, and exited the app.
Granted, its been years since i last tried Maps for gps duty, but i dont think it changed that much as far as that is concerned.

Re:Internet access for vehicle passengers (1)

mythosaz (572040) | about 6 months ago | (#47263507)

Granted, its been years since i last tried Maps for gps duty, but i dont think it changed that much as far as that is concerned.

Sure. Why bother to check when you can just reply.

Offline access - completely pre-loading towns, cities, whatever, got added in mid-2012.

Re:Everything should be done over WiFi (2)

Sockatume (732728) | about 6 months ago | (#47254881)

When they come up with a version of wifi that works over the dozens of miles that currently exist between me and my nearest tower, sure. But I suspect it would either look like a giant power plant attached to an enormous antenna, or just be a reinvention of the current cellular standards.

Re:Everything should be done over WiFi (0)

Anonymous Coward | about 6 months ago | (#47254977)

you don't need an enormous antenna, you need a groundplane large enough to cancel enough noise to pick out the signal from a distant source. Enter, stage left, the Cantenna. 1.21 inches of bare copper core rammed into the side of a Pringles can. That aluminium base is enough of a groundplane that you can use such a simple device to pull a two-way signal over ten to twenty miles. I think the record is just over 100 miles... oh, and it's line-of-sight, you're not doing an ionospheric bounce with a few tens of milliwatts.

Re:Everything should be done over WiFi (1)

tepples (727027) | about 6 months ago | (#47255081)

Good luck keeping a Cantenna aimed correctly while traveling down a road.

Re:Everything should be done over WiFi (0)

Anonymous Coward | about 6 months ago | (#47256319)

Keeping the cantenna pointed correctly is a piece of cake. Keeping the car properly pointed at the same time is what I find challenging.

Re:Everything should be done over WiFi (1)

NatasRevol (731260) | about 6 months ago | (#47255229)

Awesome.

For when you want to do wifi for only one person, who is standing still.

I'm sure it'll catch on in mobile uses real soon now.

Re:Everything should be done over WiFi (1)

TechyImmigrant (175943) | about 6 months ago | (#47255065)

>When they come up with a version of wifi that works over the dozens of miles

When people stop telling people to turn on the 'security' on their wifi and instead suggest they all share and benefit from the widespread availability of open wifi.

Wi-Fi from the street (1)

tepples (727027) | about 6 months ago | (#47255089)

Good luck keeping up with handoff from one AP to another at 60 km/h.

Re:Wi-Fi from the street (1)

NatasRevol (731260) | about 6 months ago | (#47255239)

No idea about this.

What's the difference between cell handoff and wifi handoff?

Re:Wi-Fi from the street (1)

tepples (727027) | about 6 months ago | (#47255325)

Cellular cells can be far bigger than Wi-Fi cells because cellular cells are on licensed spectrum.

Re:Wi-Fi from the street (1)

NatasRevol (731260) | about 6 months ago | (#47255957)

That doesn't address the switching that I was asking about.

Smaller cells mean more handoffs (1)

tepples (727027) | about 6 months ago | (#47256091)

Then let me explain the ramifications of cell size in the language of a geometry textbook: You have region A using technology A and region B using technology B, and technology A's cells have a much larger area. Two vehicles travel at a steady velocity, one through each region. Which will encounter many more cells? The vehicle in B. Which will have to perform many more handoffs? The vehicle in B. Now let's substitute names into this formula: Technology A is licensed cellular, and B is Wi-Fi.

Re:Smaller cells mean more handoffs (1)

NatasRevol (731260) | about 6 months ago | (#47256441)

Still doesn't have anything to do with the mechanics of the handoffs.

And these handoffs currently aren't fast enough (1)

tepples (727027) | about 6 months ago | (#47256939)

Let's just say that in practice, when I use my Dell laptop or ASUS tablet while riding the city bus, associations to known open networks take so long that by the time the IP address would normally be obtained, the bus has already moved me out of range. Nor do Wi-Fi APs serving different buildings tend to cooperate in any way, especially if they're on different ISPs (such as the local cable company, the local DSL company, or the local FTTP company). Finally, most open APs run by recognized businesses implement a captive portal, requiring me to first visit a non-HSTS web site to get hijacked and then click through a user agreement before doing anything useful. This requires manual intervention at each handoff. I don't know what if any steps the IEEE 802.11 committee plans to take in the future to make associations more rapid; we'll have to wait to see.

Re:Wi-Fi from the street (1)

TechyImmigrant (175943) | about 6 months ago | (#47255343)

Bigger cells. Coordination between those cells.

Handoff between APs on the same ESS is smoother, cleaner and quicker than cell handoff. It's all layer 2.
Handoff between APs on unrelated APs is a case of starting fresh each time.
You don't notice the first. You certainly notice the second.

Cell handoff is somewhere in between. A bunch of protocol occurs to prep the new cell and the network is using mIP so you see the same IP. It's slow and clunky, but good enough for voice and asynchronous data.

Re:Wi-Fi from the street (1)

umghhh (965931) | about 6 months ago | (#47256257)

first works, the other other does not.

Re:Everything should be done over WiFi (0)

Anonymous Coward | about 6 months ago | (#47255461)

We've used this to successfully transfer fast internet to another site. IIRC around 200ish bucks per site, check out amazon i guess
http://www.ubnt.com/

Re:Everything should be done over WiFi (1)

lars_stefan_axelsson (236283) | about 5 months ago | (#47270427)

When they come up with a version of wifi that works over the dozens of miles that currently exist between me and my nearest tower, sure.

Range is one thing, and probably the easiest to fix (compare the failed WiMax attempt). However, there are other requirements that we make of the cellular system that WiFi also ignores, such as (off the top of my head): Hand over between cells in an orderly manner, service guarantees for voice calls, emergency service guarantees (even kicking out already ongoing call of lower priority), keeping track of where the mobile is in case of incoming calls, being able to do all this with mobiles that travel at high speed etc. etc.

When WiMax tried to solve the same problems, lo and behold, it didn't turn out any cheaper and simpler than the mobile systems and hence (as many of us predicted) it failed pretty miserably. Moving into the field dominated by telecoms it turned out that they already knew about how to do that well, and the WiMax people couldn't catch up (at least not cost effectively). The evolved 3G and 4G standards that were already in place, turned out to be much more successful.

All ur data are belong to China (0)

Noah Haders (3621429) | about 6 months ago | (#47254885)

Huawei's involved? Good to know, so PRC can read my junk. At least the NSA will be frozen out.

Re:All ur data are belong to China (0)

Anonymous Coward | about 6 months ago | (#47254981)

Huawei are a privately-owned enterprise. While they may co-operate with gov.cn, they probably don't have a direct line and don't bend over backwards if they don't have to. ZTE on the other hand... they operate from the PM's office :)

Re:All ur data are belong to China (0)

Anonymous Coward | about 6 months ago | (#47255125)

but does it work on OpenSUSE?

How? (0)

Anonymous Coward | about 6 months ago | (#47254993)

The article is remarkably content-free. Is this some combination of TD-LTE and GSM with alternating timeslots? Or something more subtle?

Re:How? (1)

Hal_Porter (817932) | about 6 months ago | (#47257813)

It seems like it's based on dynamically allocating spectrum between GSM and LTE

http://www.networkworld.com/ar... [networkworld.com]

However, using a technology called GL DSS (GSM-LTE Dynamic Spectrum Sharing) Vodafone and Huawei have shown a way to allow GSM and LTE to coexist.

In a traditional mobile network, operators allocate each technology an exclusive set of frequencies. For example, many operators, including Vodafone, currently hold 20MHz of spectrum at 1.8GHz, of which 10MHz is used for LTE and the rest for GSM traffic.

GL DSS lets Huawei's SRC (Single Radio Controller) give GSM a higher priority during periods of heavy traffic, ensuring that voice calls get though unharmed. But the SRC can also provide more room for LTE when users aren't making calls, allowing for better throughput, the vendor said on Tuesday.

There's a paper on it (or at least a similar idea) here

http://arxiv.org/pdf/1302.0320... [arxiv.org]

It's interesting because it seems like GSM will live on for low bandwidth machine to machine applications even though most of the spectrum has been converted to LTE. So if you've got an embedded system with a GSM modem, there's no need to worry that the carriers will cut off the signal in order to get more LTE bandwidth.

Re:How? (0)

Anonymous Coward | about 6 months ago | (#47260047)

I believe it wouldn't work on TDD, only FDD as GSM uses FDD. 3GPP also decided to use a similar bandplan to allow a gradual (but permanent) switchover from GSM to LTE, but this technology may allow the two to coexist for as long as LTE exists. GSM uses a combination of FDMA (the ARFCN, which tells which frequency you're using) and TDMA (timeslots within that ARFCN). The trouble with GSM is that channels are pretty narrow and you can only use one channel at once, so even using multiple timeslots, there's no way to make GPRS/EDGE really fast. Capacity for more users was added by adding more ARFCNs per cell or sector or subdividing cells to make them smaller.

LTE uses OFDMA which divides it's resources in a more flexible way. OFDMA is by nature a mix of FDMA and TDMA, so the two can work together if you introduce some mechanism to transfer frequency resources from one network to the other. Another option would be to extend the baseband modem to generate both the LTE and GSM signals in the same equipment. This isn't hard as LTE already allows for non-continuous spectrum allocation. You could just turn off a few ARFCNs in cell or sector if the traffic is low and let LTE schedule users on those resource elements. When GSM traffic starts rising again, move the LTE users to other resource elements and turn on the ARFCN again.

It may or may not be possible to operate both at the same frequency as well. Careful synchronization in theory could make this possible, but timing constraints in the protocol design may preclude it. I recall that GSM timeslot's uplink and downlink are offset not only in frequency, but in time as well to avoid non-linearities in analog circuits on the phone letting the uplink interfere with the downlink. I'm not sure LTE has something like that (LTE's uplink is more flexible too, so I doubt it, but I haven't studied this enough to be sure yet) but if it has and the timings are different, then it would be very hard, if not impossible, to schedule transmissions for users of both LTE and GSM networks on the same (frequency domain) channel. But this is not a big deal as you can have significant gains without doing this.

Note that GSM already survived a few air interface changes. It was a circuit switched protocol only, until GPRS introduced a packet switched mode, allowing for flexible scheduling so multiple users could share a few timeslots and a user could use multiple timeslots too. EDGE introduced a different modulation mode, but other users just don't care as the base station schedules packet data and circuit data as to not interfere with each other. LTE is more flexible than GSM and UMTS in so many ways I would be really surprised if some technical constraint made it impossible to do this as well to increase network efficiency.

right behind the other story (1)

cellocgw (617879) | about 6 months ago | (#47255105)

Speaking of acronyms [slashdot.org] ...

I see what /. did there. (probably unintentionally)

And all these technology advances... (1)

Torp (199297) | about 6 months ago | (#47255185)

... will help me get a cheap unlimited cellular data plan how?

Re:And all these technology advances... (0)

Anonymous Coward | about 6 months ago | (#47255513)

depends. How do you define "cheap" and "unlimited"?

Re:And all these technology advances... (1)

Torp (199297) | about 6 months ago | (#47260295)

Not like the carriers currently define it, that's fo' sure.

mods 3own (-1)

Anonymous Coward | about 6 months ago | (#47257173)

Compressed audio? (1)

Neil Boekend (1854906) | about 6 months ago | (#47260625)

Will this finally include the ability to use compressed audio to make more efficient use of the GSM spectrum (over LTE of course)? We are wasting a lot of bandwith there with ancient uncompressed audio, while modern phones are easily capable of doing that properly.

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?