Original URL: https://www.theregister.com/2013/05/24/lte_wont_go_critical_until_2018_at_best/

4G LTE: Good for tweets and watching Dr Who. Crap at saving lives

Why cops, medics must stick to walkie talkies

By Bill Ray

Posted in Networks, 24th May 2013 10:44 GMT

Critical Communications World High-speed mobile broadband standard LTE, the preferred 4G technology around the world, isn't good enough for critical networks and won't be up to scratch until at least 2018.

That's according to the TETRA + Critical Communications Association (TCCA) which promotes the development of communications tech and has been lobbying 3GPP, the custodian of LTE. TCCA now has a couple of teams pushing the standard into something that can be relied upon for more important things than updating Facebook and watching Dr Who.

Most emergency services use TETRA, or one of its predecessors, which offers robust voice communications and a feature list of things that LTE is lacking. But those networks can't cope with significant quantities of data so there's great interest in getting LTE up to speed as a replacement.

LTE is great for over-the-air broadband but lacks four features considered crucial if anyone is going to take it seriously as a mission-critical technology, which are due to be added to the standard over the next half decade.

The most obvious shortfall in LTE is the lack of voice. There is a standard for it but today's users are shunted into 3G as soon as a voice call is made (or not, on EE's UK network where users routinely complain that using 4G results in them missing voice calls). For police, ambulance and suchlike voice is still the killer feature, and they'd like to speak to everyone at once too.

That comes under the heading of Group Communications, and enables one person to speak to a load of other people at the same time. In the cellular world this is known as Push To Talk and has been a commercial failure outside the US. It's a key feature of TETRA and the TCCA reckons to be on track to have the same functionality built in to Release 12 of the LTE standard - expected late 2014.

Next up is the even-more-fundamental direct routing, whereby a single handset can communicate with another handset without the presence of a mobile network. In 3GPP this is called Proximity Services (ProSe) and is a pain to integrate with a cellular service that relies on centralised routing and authentication, but is again planned for the Release 12 spec.

ProSe is intended to ensure coverage in the case of complete network failure, but even a major disaster is likely to leave base stations operational even if their backhaul has nowhere to go, so Network Resilience will enable any base station to act alone in routing calls and messages between such parts of the network operational.

That's a stark contrast to how mobile networks operate. There is some decentralisation these days, single points of failure are frowned upon even in consumer networks, but eradicating them all won't be possible until Release 13, scheduled for publication around the end of 2016.

Bearing in mind that today's 4G deployments are based around Release 8, published in 2008, with a bit of Release 9 or 10 layered in top for the advertising people ("LTE Advanced" is Release 10) it should be clear that deployment of mission-critical LTE is still a fair way off.

It needs a bunch of other stuff too, such as the ability to interrupt a call with a high-priority message, or disconnect a low-priority user to make space for someone more important, but perhaps more interesting is the lack of standards which might permit sharing a network between critical, and none-critical, users.

Your common-or-garden GSM mobile phone network has a priority system, but it lacks granularity - a switch is flicked and the proles (anyone without a suitable SIM card) gets kicked off the network. It works (IPCom patents permitting) but it's inflexible.

LTE has a hierarchical priority system, though it's not really being used and the TETRA crowd claim it's insufficient as devices requesting connections can themselves overwhelm the network before their importance has been established.

One might ask why the TCCA isn't lobbying for more granular control and lightweight authentication, to which the answer is that the TCCA, just like the rest of the critical communications industry, is still planning to get exclusive radio spectrum allocated by central governments, into which they can drop their extended LTE networks without worrying about the proles. ®