Application Aware OS’es – Knowing When To Drop Back to GPRS

Here's a thought about reversing the concept of bearer aware applications: What if the OS of a mobile device would monitor the traffic behavior of applications running on a mobile device and then reselect to the most suitable access network for the application mix currently used?

Let me give you a practical example: While only my e-mail client is checking my inbox every couple of minutes it would be much better if the mobile was using the 2G GPRS/EDGE network instead of a 3G network in order to conserve power and to improve the autonomy time. The problem is actually very real and has to do with 3G networks keeping the mobile in a much more power consuming state after each data transfer until it goes back into idle mode. For details see here.

So when the OS detects that only such periodic and small data transfers are ongoing it could attach to the GPRS network. This could work even if other applications, such as the web browser, still run in the background but aren't actively being used at the moment. If I then use the browser again to go to a new web page, the OS could then quickly go back to the 3G network.

The beauty of the approach would be that applications would not have to be modified. A little hook in the IP stack to monitor incoming and outgoing traffic and making a decision on that to reselect to a different access network, while keeping the IP address, shouldn't be to difficult to get into an OS. Maybe something for an Android programmer? Or will Apple or the Symbian foundation be faster?

Other alternatives to make mobile Internet access for background applications more power effective could be the package of Continuous Packet Connectivity features of HSPA+ (see here, here and here) or the new air interface of LTE. But if they are as efficient as GPRS/EDGE and how long it takes for them to become available remains to be seen.

Verizon, LTE and Over the Air SIM card provisioning

Verizon recently announced Gemalto and G&D as their partner for SIM cards and remote provisioning for their LTE rollout. Remote provisioning of SIM cards (e.g. change the list of preferred network operators, network name, etc.) has become pretty much common over the past couple of years but there might be a twist with Verizon and LTE:

Today, remote provisioning of SIM cards is done via SMS. When such special SMS messages are received by the mobile device it automatically forwards them to the SIM card for execution without interaction with the user. What I am not quite sure is how that will work over LTE, because SMS over LTE is not standardized. Of course it would be possible to use the "3GPP CS fallback" feature to 2G GSM or 3G UMTS to receive the SMS messages. However, in Verizon's case that might not be possible for two reasons:

  • Their legacy system is based on CDMA, which does not have SIM cards. Hence, the CDMA part of a mobile phone might not have the necessary standardized software to forward those data SMSes to the SIM card.
  • The current LTE specs of Verizon say nothing that LTE terminals have to have a CDMA part.

So I am not quite sure how over the air provisioning will work in practice in Verizon's case!? Has there been something standardized in LTE or CDMA for "native" remote provisioning of SIM cards? If you have more info, I'd be happy to hear from you!

LTE Advanced and Cooperative Network MIMO

In today's 3G networks, voice calls are often in the so called “soft handover state”, which means that the radio network controller sends and receives the data for a voice call to and from several base stations simultaneously. The mobile then receives the voice call data stream from all three cells simultaneously and combines the received signals. While this wastes some bandwidth on the backhaul, these soft handovers often help to reach mobiles at the cell edge, which receive one or more cells with a similar strength.

For HSPA data transmissions, however, the soft handover is not used as the base stations autonomously decide when to schedule data on the air interface and thus, it is pretty difficult to synchronize the transmission over the air interface of several cells. In LTE, such a direct cooperation between cells is also not foreseen for the moment, even though there might be some benefits for mobiles at the cell edge. For LTE Advanced, however, people seem to have started thinking about it again. Instead of sending the same data stream over two or more cells, however, they are thinking about a cooperative MIMO scheme, i.e. each base station sends a different data stream and the mobile then analyses each data stream separately. The result would be a higher throughput for that mobile.

As Moray Rumney points out in his recent book, though, such a cooperative MIMO scheme would be quite challenging to implement in the network. First, it would put quite a demand on backhaul capacity and second, data would have to be exchanged between the base station with a delay of a millisecond or less. O.k., it is still some years away and technology advances, but I tend to agree with him, that's quite challenging to do. What do you think?

Are SMS and Mobile Applications the End of the Clamshell?

Lately I get the impression fewer people use clamshell phones than in the past. Even in the US it seems to me that when I was last there I saw more people with Blackberries running around and happily typing away on the keyboard than people still using a clamshell phone. So why? Could it be that the clamshell format is great for voice calls, i.e. quite sexy to open it up, but does not lend itself very well to text or web based applications? The smaller screen of the clamshell vs. sliders and the act of opening up the phone to check for messages, send a quick sms, e-mail or surf the web might just be that tiny little bit too much these days!? What do you think, what are your observations?

My Mobiles

Recently, I had an email conversation with a reader and he sent me a list of mobile phones he had in the past. Interesting memories coming up. I doubt my list is equally long but I think it is time to make my own so I won't forget. Here we go:

  • I started with a Bosch 738 (I am German, the French will get the irony)
  • A Siemens S25 (my first color phone)
  • A Ericsson T39 (my first GPRS phone (!), could only do static assignment, only worked in one network)
  • A Siemens S45i (GPRS worked well after a couple of software upgrades)
  • A Siemens S55 (browser, color screen, a couple of hundred KB of flash memory with file system!)
  • A Sony-Ericsson V800 (my first UMTS phone)
  • A Nokia 6630 (only used as a modem at the time)
  • A Nokia 6680 (First phone on which I discovered the power of S60)
  • A Nokia N70 (Natural evolution from the 6680)
  • A Nokia N93 (Bought because of the camera and the twist) 
  • A Motorola V3xx (initially used as an HSPA modem, today also as backup phone)
  • A Nokia N800 Internet tablet (o.k. not quite a mobile phone)
  • A Nokia N95 (my main phone today in 2009, bought because, well, because of everything 🙂
  • A Nokia N82
  • A Nokia 5000 (low end with color screen, used as phone and to test apps on an entry level platform)

And in addition:

  • I had a Palm 3, and two HP PDA's, discontinued to use them when I bought the N70, as I could do all the calendar stuff, etc. on the mobile phone
  • A Sierra Wireless 850 1.8 MBit/s cat 12 HSDPA PC-card modem
  • 2x Huawei E220 3G USB dongles (3.6 MBit/s cat 6 HSDPA) for Internet access. In use today. They might be a bit clunky by today's standards, but they just work in all networks I have used so far, which can't be said of other 3G sticks I tried.

There we go, the list is already longer than I thought 🙂

The Digital Dividend, LTE, HSPA, TV Stations and Wireless Microphones

Over the past days there have been reports in the press that Vodafone Germany will trial LTE in the 800 MHz digital dividend band to collect experience how this frequency band can be used for rural broadband Internet coverage. The reports also mentioned that a TV station (the WDR) will also take part in the trial. A TV station?

Why, I wondered at first, as all but Unstrung did not go into the details of why a TV station is part of the trial. So, according to Unstrung, the TV station is part of the trial as they are interested, or a bit worried, that using spectrum in the 800 MHz band for broadband Internet could have an impact on their TV broadcasts in an adjacent band. Further, wireless microphones are using the digital dividend band today and it seems it is not quite clear of what can be done about that in the future (see here and here).

So the outcome of this part of the trial will be quite important to figure out if or how all of the 72 MHz foreseen to be assigned in the 800 MHz band for mobile broadband systems can be used without generating too much interference for TV broadcasts. An interesting topic to follow.

Crop Insurance in Kenya – Via SMS

And another follow up post on the University of Oxford's Future Technology Conference: During on of the many "socializing breaks" I met Eric Seuret of 3SMobile and he told me about the crop insurance via SMS application for farmers in Kenya he has developed for Syngenta, a large global agriculture company. The concept is so elegant, simple and has the potential to improve the lives of many people, so I asked Eric to send me some more details about how the service works. Here are the details:

The weather insurance solution via SMS helps protect small holder farmers against the risk of drought.  When a farmer purchases certain Syngenta seeds and crop protection products he obtains special insurance stickers.  When he has collected a certain number of stickers he gets an ID number which entitles him to 1 acre of free weather insurance for that season. To obtain the insurance he does not need to fill out any forms or contracts.  All he needs need to do is SMS a unique insurance code ID to a local number. The system then verifies that the insurance code is valid and confirms the farmer's insurance status.  If the farmer has a larger plot of land he can obtain multiple cards and insurance policies.

Throughout the course of the year the farmer is kept informed of major weather issues that are related to the policy via SMS.  At the end of the season the historical weather data is used to determine which farmers are entitled to a payout based on their region.  In the case of a payout the farmer is notified via SMS and they can go to the stockist to obtain the payout.

It is the first time that insurance has been made available to small holder farmers in Kenya with 1 acre or less of land.  This has only been possible through the use of mobile technologies were the solution finally became cost effective.  Mobile phones were used to simplify the administration, reduce costs, enable immediate registrations, and to help form a direct link to the end customer. In many cases the mobile phone represent the only way to reach the farmers without physically visiting the remote villages.  The innovative solution is currently running as a pilot project and is so far proving to be very successful.

The insurance concept and solution was designed by Rose Goslinga and Fritz Brugger for the Syngenta Foundation for Sustainable Development and developed by Eric Seuret.

Early Beta of Nokia Point & Find

Except for Japan, 2D barcodes haven't really made it into the consumer world so far. Maybe we can jump right to the next step in the evolution with Nokia's Point & Find early prototype that does not rely on a barcode to identify something and then get some more information from the internet but instead does image recognition? Below is a video of how it works in the context of getting more information about a movie by pointing the camera phone at a movie poster.

Personally, I can also imagine to use the application for sightseeing. No more tracing through a tourist guide, just point your phone at that church, castle, monument, etc., and you get background information on it. I'd really like that.

The web page for the new application contains some further information in PDF files but doesn't go into the details of exactly how it works, like for example if a picture is taken and uploaded for analysis or if  information is downloaded before and then analyzed locally).

In any case, I think there's great potential behind this, especially because individual companies can create their own Point & Find worlds. If Nokia allows some sort of billing for world downloads, I can imagine it might be an interesting application for companies working on tourist guides (to come back to my example above) to create an enhanced electronic version of their printed guides. Free point and find world downloads would be good for other purposes, like the movie industry, who could promote their movies this way in the hope users buy a ticket online or later on rather than charging for general use.

What happened to 3GPP’s Earthquake and Tsunami Warning System?

Over the weekend, I wanted to take a closer look at the 3GPP Earthquake and Tsunami Warning System (ETWS) for GSM, UTMS and LTE. I was expecting to find proper stage 1, 2 and 3 documents in place by now so implementations could follow soon. It turns that of the three documents, only stage 1 (3GPP TS 22.168) has made it so far into Release 8. However, it seems that the specification might have been withdrawn at SP-43 in Biarritz in March 2009. 

The same applies for the stage 2 documentation 3GPP TS 23.168, it's current state is "withdrawn". There is a pointer to Technical Report (not a specification!) TR 23.828 which gives an overview of the different options analyzed and the proposal at the end to go for solution 5. Solution 5 would be to send a first warning to the mobiles via a special paging message with further details right afterwards in a cell broadcast message. For GSM mobiles already engaged in a voice call, the BTS would create a special SMS and sent it over the slow associated signaling channel while the call is in progress. In UMTS, paging messages can also be received during a call. What's not mentioned is how mobiles currently having a GPRS TBF established would receive the notification as there is no associated circuit switched signaling channel and pagings would not be received over the paging channel. But for the moment, that is beside the pont, since both stage 1 and stage 2 of the specification was withdrawn.

However, there is a link to 3GPP TS 22.268, a stage 1 specification for a much more generalized public warning system (PWS). No stage 2 specification exists so far, so this one is also at the beginning. It's an interesting document as it shows how much time there is to warn the public about an earthquake or a tsunami. Very challenging!

I am a bit puzzled!? Looks like ETWS was already quite well on it's way but now seems to be suddenly stopped? Did it fall short of the requirements? If you know more, please leave a comment below.