* Posts by ulbdd

9 posts • joined 17 Feb 2011

British government has bought a £200m 5G 'academic wet dream'


It's not a typo, and it makes perfect sense once you know the details. There are 3 variants of 5G (massive IoT, URLLC for ultra-reliable and low latency, and massive broadband for high speed) and here it's the massive broadband one that is relevant. It is using high frequencies, sometimes "only" at 5 GHz but the main interest is above 10 GHz with "millimeter waves". At those high frequencies, the range is very short : about 200 to 300m, and to even get there one need a lot of antennas for beamforming (to focus the signal on the target).

In the end, mmWaves will also be used for mobile. But using it for fixed access first makes perfect sense : the cost of fiber is mostly on the last leg of the connection, and this can be handled by 5G in a cheaper way. In the US, it will enable the telcos to challenge the cablos. Also, it's easier to start with fixed access: when doing aggressive beamforming, the signal is tightly focused on the target and nothing else. If the beam is not properly directed, one loose the connection. As you can imagine, it can be an interesting problem for mobile applications, where there is often no line of sight but the radio beam bounces before reaching the smartphone. A quick turn of the corner is a fast and brutal change. For a fixed connection, this issue disappear: the beam needs to be tuned, but it's very slow changing and easy to do. Plus for fixed access the constraints on size and power consumption are of course much relaxed compared to a handset.

So in summary: there is a business case for fixed access with 5G, and it's also an easier first step before full mobility.

Every LTE call, text, can be intercepted, blacked out, hacker finds


Clarification on LTE call/SMS interception (not!)

When they say the attack allows intercepting calls and SMS: this is only happening when the device is on 2G, not on LTE. It is still NOT possible to do the interception on LTE itself, so the attack switch the device to 2G, which is insecure.

To give the history here: 2G has no mutual authentication. So a rogue 2G base station can do MITM and intercept call and traffic. 2G has deployed often has weak crypto too (there's a fix, but not always deployed). So 2G has poor security, and tends not to be upgraded for cost reasons. But several operators are going to (or already have in SK and Japan) turned 2G off. Getting rid of 2G is the best solution here. But in Europe we'll have to be patient...

About LTE now. The initial messages between a device (UE) and base station (eNB) are not encrypted. Pretty normal there, one need to establish a context. The redirection the attack is using happens in this non-encrypted phase, so can redirect the UE to a fake (no service) or 2G (MITM) base station. There's a trade-off here: for overload management, a fast redirection is better. For security, waiting after authentication will be better (but would load the chosen cell). Pick your poison...

In practice, with pure LTE, the redirection attack is a form of DoS. And anybody who knows radio knows that jamming is easy anyway. Instead of faking an eNB, just jam the channel and kill all LTE on the given frequency. So preferring load robustness in this context is a reasonable trade-off. It's just a poor fit with network still using legacy 2G with crappy security unfortunately.

So let's get rid of 2G fast, please.

British unis mull offshore EU campuses in post-Brexit vote panic


Funny, it's quite common to hear jokes about UK plumbing in France. In particular anemic water pressure in showers. Maybe it's a case of "filer à l'anglaise" vs. "taking a French leave"?

Nul points: PM May's post-Brexit EU immigration options


Strange logic

"Offers like this for EU citizens could even help the negotiations over Brexit: the UK wants some access to the EU single market for the likes of financial services, so in return it could offer EU nationals better access to British jobs".

I have a hard time making sense of this. If we talk about highly qualified people, the big European countries certainly do NOT want those people to work in the UK. They would prefer to see them stay at home, generate business and pay local taxes. There's no unemployment problem there. Does anyone see Germany or France say: "hey, the finance jobs we want to take away from you? You can keep them, and on top we're happy to send you our best qualified people."?

And for low qualification people, well the UK Brexiters don't want them in.

So I don't see a realistic match there.

'Get a VPN to defeat metadata retention' is good advice. Sometimes


Why tethered traffic doesn't go through a VPN

3GPP cellular standards (2G/3G/4G) can support several "Packet Data Network" (PDN) connections at the same time. Each PDN appears to the device as a separate IP interface. Some PDN are for administrations or specific purposes and access to them are limited to specific applications.

How many PDNs are used and for what can vary depending on the operator.

For example, a LTE network could use:

- an IMS PDN for IMS functions (SMS, VoLTE). That's standard. Access limited to IMS software, and in some case this PDN can even be terminated on the modem chip (not visible from the AP);

- an administration PDN, for device management (using OMA-DM typically);

- a PDN for Internet connectivity. That's the one for the user traffic, most of the time. That's the one PDN the VPN software will intercept.

When you enable tethering, on many operators it lead to a specific PDN being activated and used for this traffic. It's a new interface, and the VPN doesn't handle it at all. One would need the PDN to trap this new PDN creation and insert itself in the flow, but that doesn't seem to be supported.

Bottom line: you need to run a VPN on all tethered devices to be fully protected.

Samsung, Apple soap opera drama: Korean giant WILL churn out chips for iPhones, iPads


Re: 14nm versus 16nm

This is particularly true here: the only reason TSMC calls their next node 16nm and not 14 as others is because 14 is unlucky (for some reason I forgot, maybe a Chinese speaking person can comment). Everybody in the industry considers 16/14 as the same node, and if there will likely be performance variations the 14/16 number itself is meaningless.

And by the way, for TSMC it's best understood as mostly 20nm with FinFETs added. There is also a 16FF+ coming quickly after with some incremental performance and density optimizations. As I understand it's the same for Samsung/GF: the node change is justified by the move to FinFETs and the performance/power impact, not by a true shrink.

Intel pops $4.1bn to save Moore's Law from repeal


Re: Smaller still means hotter right ?

It's the opposite. The smaller the process, the highest the leakage. But then, the lowest the operating voltage. So by going to a smaller process, you reduce the dynamic power consumption (due to lower operating voltage) per transistor, at the price of increased leakage while the transistor is powered. But the leakage part can be controlled by power gating (= cutting the power) parts of the chips when they're not needed.

Nokia unveils Contractual Obligation Meego Phone


Customers will decide

Yes. Also, whatever the plans of Elop and Nokia board (and it doesn't look as their plan develop exactly as expected so far...), it will be the customers who will decide the future of Meego at Nokia.

If sufficient Nokia users decide to try the N9, it will have support. Particularly if WP7 continues to be the commercial flop it currently is. And whatever Elop preference is, because in the end it's the bottom line that counts.

Another factor there: I'm sure a lot of employees in Nokia are rooting for the N9 vs. WP7 phones and don't see them as Microsoft OEM. They'll push all they can for the N9. If there is sufficient support in front from Nokia phone users, there can be a meeting of souls and money following to turn the big shots opinion in time.

So in the crazy situation Nokia is right now the future is actually more open that what PR declarations from high on want people to believe.

A beer, as a toast, to all the Nokia people who worked on bringing what looks like a nice phone is such an adverse context. I may get one myself.

Spanish whispers on Microsoft and Nokia


About convergence

Pretty much the same feeling about a converged Meego device. That's really what I wanted.

But it seems clear to me that Nokia has convergence in mind alright, just based on Windows. Think about it:

1) MS announced support for ARM in W8. Everyone thought about low power server, but this also opens the door to using the same core for their handset offering;

2) WP7 is a new UI on top of an old (and poor) embedded OS. They want anyone (but them, when it suits them of course. That's how MS likes competition as we all know ;) to use the .Net VM as they don't want any dependency on this poor OS. Why? Because they know they'll have to change the foundation for a modern kernel as soon as they can.

So it's pretty clear to me that MS, in the time frame of W8, will be like Apple: the same foundation for phones, PC and server OS. And all .Net phone applications will be able to run unchanged in phone mode.

But then, when docked to a larger screen and keyboard, the phone will be able to be used as a computer. And support any regular PC applications, for ARM.

Yes, for ARM so all current x86 applications will need to be ported to ARM. I'm sure it's not so complex and MS will do it for Office and Outlook. And if it works, others will follow.

They could also stay with x86 (future Atoms) but I'm not sure they care. Getting rid of Intel and moving to a more dynamic ecosystem with so many CPU provider is better. A smaller part of the cake for the chip vendors, so possibly a bigger for the software and service providers. And recompiling for an ARM W8 shouldn't be too difficult.

On the other hand, if the OS changes there's little interest in keeping the same instruction set. It's Intel who gets the shaft here.

So you read it here: Nokia will help to re-establish MS monopoly in the new converged era. By using all those lazy people eager to find the Office and Outlook they know (for biz people at least). And with the trend of using the same device for personal use and work, this office addiction could help in the consumer space.

Aaaaargh! Hopefully a linux based alternative will emerge. But the champion is Android, and Android is too short for a converged solution (see the Moto Atrix, nice idea but too limited). And Google only care about getting you to their service. So will they make the effort (and spending) to offer you a proper local OS too?

I would be very depressed to see the same kind of crap in the next coming generation. But it seems that's what Nokia has in store for us.

Biting the hand that feeds IT © 1998–2019