Making a mesh of network coverage

This can lead to the frustrating situation where you know the person you want to get hold of is quite close to you – for example, you’ve lost your friend in a shopping centre – but their phone isn’t in coverage so you can’t contact them.

This is the type of scenario Swedish start-up TerraNet AB is trying to resolve with their ‘mobile mesh’ networking technique.

The ides is rather than routing calls via a local base station, each handset tries to find others that are within direct radio range (up to two kilometres away). As more mobiles come into range, they arrange themselves automatically into a mesh. Each one in the mesh is able to operate in a repeater mode to join handsets which would otherwise be too far apart to talk to each other, with intelligence to work out how to get calls from one phone to another by bouncing off intermediaries.

The current system allows up to seven intermediate handsets, so giving a theoretical maximum of 16km between handsets in a conversation. This isn’t a hard limit – TerraNet also has a USB adaptor for computers, which will allow an internet connected PC to act as a VoIP gateway device to give access to the wider telephone networks.

From a user point of view, the mesh routing happens automatically. All they have to do is dial a number, and if that handset is in direct range, or can be reached by being passed among intermediate handsets, then the call will connect.

TerraNet’s main target for this technology is developing countries, or areas which have just suffered a natural disaster. In either of these cases, the TerraNet system will work to give rudimentary communications without the presence of a traditional commercial mobile network.

It would also have applications in areas of reasonable population density where many local calls would be made – for example on a university campus. As the calls are all routed between devices on the mesh, there are no ongoing costs once you’ve purchased the hardware.

Obviously this wouldn’t be too popular with the mobile network operators, though it won’t sound their death knell. While the mesh system is technically clever (self organising networks are quite difficult to arrange properly, especially when all of the nodes are moving relative to each other so the routing has to be very dynamic), it can be very hungry for radio frequencies – especially if there are many hops being used to connect calls.

It also has a completely different quality of service to existing networks; with these you know that if you place a call in an area of reasonable signal strength and don’t move around, the call will remain connected. With a mobile mesh, the call connection depends upon enough other users being in place between you and the person you are talking to – this requires a significant take up before the system is reliable enough for general use.

Flash in the pan

The electronics industry has decided there should be yet another type of memory card.

Not content with Compact Flash, Memory Stick, XD, SD, MMC, RSMMC, microSD and the dozen other alphabet soup variants, the new wonder card will be called UFS – universal flash storage.

This is being developed by the JEDEC Solid State Technology Association, one of the leading worldwide electronics standardisation bodies, with support from memory manufacturers such as Micron, ST and TI, plus mobile companies including Nokia, Sony Ericsson and Samsung – which should guarantee reasonable uptake when it is finally completed in 2009.

The main advantage over current cards is the data transfer speed; UFS should be capable of up to two gigabits per second; a significant increase over existing cards. There will still be a limit in the write speed on the actual flash devices though this, together with the memory density, is improving with each new generation. UFS also allows newer memory technologies, such as phase change RAMs, to be used. Some of these will benefit even more from the better bandwidth, which will be especially appreciated in portable multimedia players.

Of course, this is still just another memory type to fight for consumer acceptance, and in the relatively fast moving memory world it could easily be on the way to being superseded by the time it reaches the market. Company allegiances, such as Sony Ericsson’s current desire to use the Sony Memory Stick in order to maintain compatibility with the rest of Sony’s consumer electronics range, will also be a factor.

Mobile Ads make Sense?

Mobile advertising is going to have a bit of a shake-up, with Google and Nokia entering the arena.

Google has officially launched its AdSense for Mobile product. This works in a very similar way to the existing AdSense service; publishers of mobile websites can embed the AdSense code into their page, then Google will fill that in on-the-fly with adverts which are related to the content of the page. If the reader of the page clicks through an advert, the publisher receives a payment from Google.

On the other side of the business, advertisers ‘bid’ with Google for positions on pages, with adverts related to popular keywords costing more – sometimes much more – than less popular keywords.

It remains to be seen if this will be a successful model for mobile; browsing habits are different to those of desktop users, with quick, focussed information in demand rather than random browsing.

Nokia has said it wants to branch out into consumer internet services and has bought its way into mobile advertising by buying the US-based Enpocket agency. Enpocket has expertise in mobile advertising, running SMS, MMS and WAP campaigns for some of the world’s largest companies, such as Pepsi and Ford. Between Nokia and Enpocket there is a lot of mobile experience, but will it be enough to challenge Google long-term in the advertising business? We’ll see.