Adaptive Video Streaming: Principles, Improvements and Innovation

An Interdigital presentation from last year explains the principle of adaptive streaming very well for those who would not know how it worked.

This process of adaptation could be improved based on the quality of coverage at any particular time.

Interdigital are proposing a further enhancement of improving the adaptation further based on the User behaviour. If for example the user is far away then the quality need not be great on the device. On the other hand if the user is very close-by, the quality should be as good as it can get. They have explained it in a whitepaper for whoever is interested here.

A video showing this method is embedded below:

500 Billion devices by 2030, etc...

Few weeks back in the LTE World Summit 2013, I heard someone from Ericsson mention that internally they think that by 2030 there will be 500 Billion Connected devices on the planet. The population projections for 2030 is somewhere around 8.5 Billion people worldwide. As a result the figure does not come much as a surprise to me.

John Cunliffe from Ericsson is widely credited for making the statement 50 Billion connected devices by 2020. Recently he spoke in the Cambridge Wireless and defended his forecast on the connected devices. He also provided us with the traffic exploration tool to see how the devices market would look up till 2018. Here is one of the pictures using the tool:

In terms of Cellular connectivity, we are looking at 9 Billion devices by 2018. The interesting thing to notice is that in 2017, there are still some 4 Billion feature phones. While in the developed world our focus is completely on Smartphones, its interesting to see new and existing SMS/USSD based services are still popular in the developing world. Some months back I heard about Facebook developing SMS/USSD based experience for Feature phones, I am sure that would attract a lot of users from the developing world.

One thing missing from the above is non-cellular connections which will make bulk of connectivity. Wi-Fi for example is a major connectivity medium for tablets. In fact 90% of the tablets have only WiFi connectivity. Bluetooth is another popular method of connectivity. While its mostly used in conjunction with phones, it is going to be a popular way of connecting devices in the Personal Area Network's (PAN's). So its no surprise that we will see 50 Billion connected devices but maybe not by 2020. My guess would be around 2022-23.

Is it too early to talk '5G'

While LTE/LTE-A (or 4G) is being rolled out, there is already a talk about 5G. Last week in the LTE World Summit in Amsterdam, there was a whole track on what should 5G be without much technical details. Couple of months back Samsung had announced that they have reached 5G breakthrough. In my talk back in May, I had suggested that 5G would be an evolution on the Radio Access but the core will evolve just little. Anyway, its too early to speculate what the access technology for 5G would be.

Ericsson has published a '5G' whitepaper where they talk about the vision and why and what of 5G rather than going into any technical details. It is embedded below:

5G: What is it? from Ericsson

Multi-RAT mobile backhaul for Het-Nets

Recently got another opportunity to hear from Andy Sutton, Principal Network Architect, Network Strategy, EE. His earlier presentation from our Cambridge Wireless event is here. There were many interesting bits in this presentation and some of the ones I found interesting is as follows:

Interesting to see in the above that the LTE traffic in the backhaul is separated by the QCI (QoS Class Identifiers - see here) as opposed to the 2G/3G traffic.

This is EE's implementation. As you may notice 2G and 4G use SRAN (Single RAN) while 3G is separate. As I mentioned a few times, I think 3G networks will probably be switched off before the 2G networks, mainly because there are a lot more 2G M2M devices that requires little data to be sent and not consume lots of energy (which is an issue in 3G), so this architecture may be suited well.

Finally, a practical network implementation which looks different from the text book picture and the often touted 'flat' architecture. Andy did mention that they see a ping latency of 30-50ms in the LTE network as opposed to around 100ms in the UMTS networks.

Mark Gilmour was able to prove this point practically.

Here is the complete presentation:

Timing Accuracy and Phase Performance Requirements in LTE/LTE-A/4G

Nice quick summary videos from Chronos.



If you are interested in learning more on this topic or discussions, I would recommend joining the Phase Ready Linkedin group.

3 Band Carrier Aggregation in Release-12

So it looks like in the latest 3GPP RAN meeting finally more than 2 carriers have been proposed for Carrier Aggregation. The TDoclist has a few items on 3 carriers for CA. In some cases its been specified that there is 1 uplink component carrier (1UL CC) but in other cases its not specified and I have not looked into details. Its good to finally see more than 2 carriers being discussed.

Rohde&Schwarz have explained in one of their whitepapers about the numbering of CA bands.

Now there is a possibility that we may have 2 contiguous bands and 1 band from an Inter-band so the naming would be accordingly. There are also going to be new carrier types (NCT), Band 29 for example. See details here.

Finally, If you want to learn more about Carrier Aggregation (CA) or other LTE-Advanced features, my article from last year, here, would be useful.

Economical M2M using LTE - #LTEWS

In the upcoming LTE World Summit 2013 (programme here), I will be doing a briefing on the topic 'Economical M2M using LTE'. I have some ideas but I would like to hear more on what you think? In fact, is LTE the right technology from the M2M device point of view? Or do they better stick to 2G (I dont think 3G is good enough generally from low data M2M point of view). What other issues can be foreseen? Security? Roaming?

A recent presentation from Telefonica shows how they are partnering with other operators worldwide to create universal solutions. Will this help? Why not use these solutions for everything, not just LTE? LTE is data only technology isn't it?

The presentation is embedded below to draw your own conclusion but I an interested in hearing your thoughts on Twitter or here on the blog.

Telefónica - M2M Global Operator – Strategy of Alliances from Zahid Ghadialy

Five Future Technologies (#FutureTech) we may see soon in our devices!

Remember 'Internet of Everything' rather than 'Internet of Things', here is a recent Cisco video explaining this vision:

You can read more on these topics on Cisco blogs.


Next is the self-healing technology:



Spray-on clothes
These will go very well with phones where you can spray paint phone covers and maybe if its possible to dissolve the skin and re-use it, it would be an added bonus.


Just think how the technology used to design Robots can enable flexible phones and other devices.




With sensors becoming smaller and cheaper, more of them are being put in our devices. Many years back people were saying that breath analysers could be available in mobiles but I guess there wasnt a business case for that. Also many of these sensors have come as part of Bluetooth add-ons to keep the cost/weight/size of the device down. Now there is a possibility of whole new range of sensors coming to our devices.

You can read more details on this here.


Finally, there is always one user who would ask me why is there no mention of LTE in the videos above so here is a **bonus** video.

I have to mention that this didnt sound very convincing to me as a selling point. Its like back in the year 2000, 3G was being sold as an enabler to the must have 'Video calling'.

Summary of Network Security Conference (#NetworkSecurity) 2013

Summary of Network Security Conference (#NetworkSecurity) from 3G4GLtd

3GPP Public Safety focus in Rel-12

Public Safety is still a hot topic in the standards discussion and on this blog as well. Two recent posts containing presentations have been viewed and downloaded like hotcakes. See here and here.

3GPP presented on this topic in the Critical Communications World that took place last month. The following is from the 3GPP press release:

The ’Critical Communications World’ conference, held recently in Paris, has focused largely on the case for LTE standardized equipment to bring broadband access to professional users, by meeting their high demands for reliability and resilience.

Balazs Bertenyi, the 3GPP SA Chair, reported on the latest status of the first 3GPP features for public safety, in particular those covering Proximity services (Direct mode) and Group call. He spoke of the need to strike a balance between more or less customisation, to make use of commercial products while meeting the specific requirements for Public Protection and Disaster Relief (PPDR).

To ensure that these needs are met, Balazs Bertenyi called for the wholehearted participation of the critical communications community in 3GPP groups, by sending the right people to address the technical questions and obstacles that arise during the creation of work items.

A presentation and video from that event is embedded below:

LTE Standards for Public Safety – 3GPP view from Zahid Ghadialy

For more details see here.