The Crisis in UK Critical Communications

Peter Clemons is an expert in the field of “critical communications”. This includes public safety services that we all depend upon in times of personal and national emergencies. He sees an unfolding crisis in this space.

To my mind, this reflects wider issues that affect all applications of telecoms, whether safety of life is involved or not. Are the services we are building fit for purpose? Do they represent value for money? We talked about the situation in the United Kingdom, where both safety and value appear to be at risk.

MG: What’s happening in Critical Communications right now in the UK?

PC: The ‘big picture’ is that public safety has been based on specialist Terrestrial Trunked Radio (TETRA), and is now moving to a converged 4G LTE platform.

To this end, the UK government is commissioning a new emergency services network. This will provide service for around 250,000 first responders: police, fire, rescue, and ambulance. On top of this there are around 50,000 additional “tier 2” users. These users include utilities and organisations like Transport for London.

All users currently have voice-centric TETRA handsets from Airwave. The problem is that it is not clear how that migration to LTE will go ahead in a safe and cost effective manner.

Tell me more about the UK TETRA network, since this is a specialist technology.

To understand how we have got into trouble, it helps to have a bit of history of this sector. Before TETRA there were dozens (if not hundreds) of different analogue local mobile radio networks. As a kid, I could listen in to local police and ambulance – as could criminals and gangsters.

Technically, TETRA is an ETSI standard that is similar to GSM. Whereas GSM was designed for mass market use, TETRA was tailored to emergency services work. The needs of these users are centred on mission-critical voice and group calling, with high levels of security and resilience.

Commercially, the initial contracts were signed around the year 2000. The TETRA network has intentionally been created as a private monopoly, with BT being the operator, and Motorola the network supplier.

It took about five years to deploy TETRA nationwide. The terrorist attacks in London in 2005 helped to accelerate user adoption. For example, at that time London Underground had a separate network that was not compatible, so this deficiency was rectified.

As a result, over the last 7-8 years the UK has established what is considered to be one of the more complete and functionally rich networks anywhere in the world. It has coped with floods, the Olympics, Royal weddings and more. It is often the one network left running during crises.

TETRA sounds like a great success!  So why are we putting that at risk?

In most countries, both in Europe and around the world, the government has kept control over critical communications services. The service provider tends to be a government department, or at least is accountable to government.

In the UK, however, the government decided to carry out an experiment. They contracted for service to be provided by private company, namely BT. This subdivision, BT Airwave, then became O2 Airwave when Telefonica bought O2. That put the Spanish in charge of delivering the UK emergency services network.

Telefonica decided in 2007 that this was no longer part of their core function, and sold the division to an Australian-owned investment fund, Macquarie. The UK taxpayer is now paying them £450m/year to run the network.

Meanwhile, the rest of Europe had built out their own critical communications networks, and kept control over the operation of them. As a result they have per-user prices that are around one fifth of those in the UK for similar nationwide coverage.

So in 2010, the new Coalition government in the UK turned up, and was determined to “pay down the deficit” (not that they have achieved any such thing). They reviewed government contracts and evaluated them for their efficiency and effectiveness.

They saw this huge price gap for emergency services communications provision, and considered it to be a serious problem. Surely we could find a way to provide a similar service for a lower cost?

Ouch! That sounds like good news for contract lawyers. What happened next?

What has followed has been years of wrangling between the Airwave hierarchy and the UK Treasury and Home Office. The network operator is focused on shareholder value, and hence is keen on enforcing the terms of the original agreement. The government is aiming to the reduce cost of that contract towards current international norms.

The trigger event has been the auction of 800Mhz and 2.6Ghz spectrum for 4G LTE. This technology has been promoted as “does everything for everyone”. LTE has low latency, high data rates, and megahertz of spectrum (compared to kilohertz for TETRA).

The Treasury has assumed that it is technically possible to switch the emergency services over to a commercial LTE network. After all, TETRA was specified in the 1990s for narrowband voice with very limited data.

Meanwhile, mobile network operators have paid billions of pounds for prime spectrum. Critical communications represents a new vertical, with the opportunity to test new business models. There’s just one problem: how can they divide their network resources up to reflect these very different priority needs?

So why should people be concerned?

Things are not as technically simple as they seem, and as a result the public should be deeply concerned, both in their roles as citizens as well as taxpayers. There is a risk of a safety impact on ordinary people, and a cost for everyone. Why so? LTE was not designed for critical communications.

Commercial mobile operators have moved from GSM to UMTS to WCDMA networks to reflect the strong growth in demand for mobile data services. Smartphones are now used for social media and streaming video. LTE technology fulfils a need to supply cheap mass market data communications.

So LTE is a data service at heart, and reflects the consumer and enterprise market shift from being predominantly voice-centric to data-centric. In this wireless data world you can still control quality to a degree. So with OFDM-A modulation we have reduced latency. We have improved how we allocate different resource blocks to different uses.

The marketing story is that we should be able to allocate dedicated resources to emergency services, so we can assure voice communications and group calling even when the network is stressed. Unfortunately, this is not the case. Even the 3GPP standards bodies and mobile operators have recognised that there are serious technology limitations.

This means they face a reputational risk in delivering a like-for-like mission-critical voice service.

Won’t this be fixed by updated standards?

The TETRA Critical Communications association (TCCA) began to engage with the 3GPP standards process in 2012. 3GPP then reached out to peers in the USA and elsewhere: the ESMCP project here in the UK, the US FirstNet programme, and the various European associations.

These lobbied 3GPP for capabilities specifically aimed at critical communications requirements. At the Edinburgh meeting in September 2014, 3GPP set up the SA6specification group, the first new group in a decade.

The hope is that by taking the critical communications requirement into a separate stream, it will no longer hold up the mass market release 12 LTE standard. Even with six meetings a year, this SA6 process will be a long one. By the end of the second meeting it had (as might be expected) only got as far as electing the chairman.

It will take time to scope out what can be achieved, and develop the critical communications functionality. For many players in the 3GPP process this is not a priority, since they are focusing solely on mass market commercial applications.

OK, that’s not going to be a quick fix. What might be the consequences for the UK public if we keep on the current course?

We now have a multi-billion pound government contract that will be awarded during summer of 2015. The contract has already been drafted. Now the elections are over, the new Conservative government will press forward. The intention will be to replace TETRA with commercial LTE, with no new spectrum dedicated to the service.

The supplier selection process has been reduced to just two bidders: O2 and EE. These are both is in the process of being acquired, by Hutchinson and BT respectively.

Yet there is no LTE public safety technical standard, nor will there be one for a long time. Consider a standard voice services like VoLTE, which has been in development and deployment for years. At this moment in the US a Verizon customer can’t yet communicate with an AT&T one over VoLTE. We are forced into circuit-based fall back solutions.

So there is de facto no standard, even for traditional voice communication, let alone for mission-critical voice.

How might this cause problems for users?

Well, walk down any High Street in the UK, and you will see police officers who wear a TETRA radio attached to some part of their body.

The radio has a big red button, so if they are chasing a criminal they can press the red button, and his or her communications go direct to a control room. It takes milliseconds to establish their location, and seconds to dispatch support resources.

This is not available as a standard over any commercial mobile network. It’s a basic safety issue: there is no guarantee emergency services personnel will have the same functionality as under TETRA. Yet every single user survey says the top requirement is mission critical voice. Data, maps, and media are all secondary.

That means we are left with workarounds for emergency voice services, which raise many questions of technical and economic feasibility. For instance, suppose we develop custom LTE handsets. What is the cost? Will they interoperate? Are we now dependent on one Chinese manufacturer?

If we are not careful, we will find ourselves in the same situation as GSM-R. This GSM variant is a standard adapted for use on railways. It comes in just one version, with limited suppliers. The worst case scenario we now face is a fiasco: a service that is not as safe as TETRA, yet is even more expensive.

Why did we get into this situation?

There are many problems of process, values and technology.

There is a lack of transparency to the bidding process. The contract is being negotiated secretly, and everyone is under NDA. They can only discuss the Ts&Cs once the contract is signed. This poses a danger to emergency services personnel, the public, and the taxpayer.

The government contract is for basic network service provision only. The police will have to purchase terminals and supporting services, as these are not included. They are very worried, as this involves lots of risky new technology development. Indeed, it is possible that we may have to wait for 5G (whatever it is) to replace 2G TETRA services.

The interim solution may be a 3G or LTE type data service that is provided as an overlay on TETRA. This may be done through MVNO contracts. However, there is a danger of early technical obsolescence of any hybrid solution.

The underlying technical issue is one of safety critical systems engineering. We see hope triumphing over engineering. This optimism bias is unsuitable for emergency services.

For example, with Airwave there are two totally redundant and resilient control rooms in secret locations. As we move to a commercial LTE network model, we may find by chance that networks have been built that are as resilient as public ones, but nobody can know for sure.

With TETRA, at least ⅓ to ½ of the generators for backup power need to last a week. When commercial networks lose power, they go down quickly. They are designed for providing best effort data to tens of millions of users. We’re trying to deliver two different models with one network, and it’s not working.

What would you like to see happen next?

I would like to see the new government take a deep breath, step back, and acknowledge their strategic misstep. What the rest of Europe is doing is a better template to follow. (The US will take decades to complete if they continue at their current pace.)

It’s not a black-or-white world where we need to get rid of the old network and build a new one. We don’t have to restrict ourselves to just two options of “switch off TETRA” or “allow TETRA”. Commercial LTE is not fit for purpose for critical communications, so it is in nobody’s interest to deploy it as a sole solution.

We should allow the 3GPP standards process to take its natural course. This will take 3-5 years. In the meantime, the Treasury is not happy with Airwave. The new government should seize this opportunity to engage with them in a more positive way. They do they want to continue running the TETRA network. A new contract needs to be agreed at a lower cost than the current contract.

Meanwhile, the government should continue its discussion with the current bidders. There should be a 12-18 months hiatus to allow the acquisitions of O2 and EE to be completed. By this time we will be in a better place to exploit the true coverage and capacity benefits of commercial LTE. In the meantime we can conduct more pilot trials.

This process needs to happen in a more open and transparent way. By failing to engage properly we have created enormous technical and commercial risk. Any future contracting entity needs to be spun-off from the commercial operators and tightly controlled by the public sector.

This all requires some brave decisions, and compromises that are acceptable to the bidders for the UK’s critical communications needs. If we do the right thing, the UK would then move into position more in line with rest of Europe. That matters, because we’re in a digital single market.

The ability of UK companies to compete elsewhere is being limited, as their hands are tied locally. The rest of Europe doesn’t believe the UK is fully committed, and this is a global game. There are already countries like Qatar that are leapfrogging us with a commercial version of LTE. It provide services that TETRA can’t do, but without ditching TETRA.

To learn more, visit the TETRA applications blog. To get in touch with Peter, you can email him at peter@quixoticity.com.