'Inherent risk' to untried and untested 4G emergency services network – NAO

Plans for a new 4G emergency services network (ESN) to be used in life and death situations by the blue light services have been dubbed as "high risk," in a report from the National Audit Office today. The ESN will replace the expensive Airwave radio system used by the police, fire and ambulance services. It is expected to …

[Anonymous Coward]

Good luck on new years or if there's a major emergency, I seem to remember that when planes hit vital parts of New York's infrastructure there wasn't much in the way of mobile signal.

And the bit where it goes "The closest is South Korea" it should of said "but they're doing it properly"

So do the costs include the blue light services needing their own portable "towers" for when disaster strikes and the impact incidents would have on the national bottom line and lives while the 4g signal is down?

[Dan 55]

I don't know why it's taken the NAO to say this, and at this late stage. Has everyone else in the civil service been lobotomised?

It seems that for any project you care to mention, if almost nobody else in the world does something in this way then it's like a red-rag to a bull and brinks out PHB's that use phrases like groundbreaking and cutting-edge technology and so on.

[Anonymous Coward]

This sounds more like a ministers idea than civil service...

Civil Service never care about saving money, its the Minister wanting to modernise and save without having a Sir Humphrey to steer them correctly

[Anonymous Coward]

So doing something first is considered a bad thing? Not going to argue that there's risk, but as I understand it the functionality is there in the Standards to handle all the issues re loading, user prioritisation, etc. Time will tell.....

[Charles 9]

"So doing something first is considered a bad thing?"

YES, when it takes priority over doing it RIGHT.

[NotBob]

So do the costs include the blue light services needing their own portable "towers" for when disaster strikes and the impact incidents would have on the national bottom line and lives while the 4g signal is down?

One assumes, perhaps incorrectly, that it at least includes making emergency services' devices take priority in heavy congestion. The tech exists. Of course, they may instead be relying on some sort of "cell on wheels" or "cell on light truck" portable tower solution. Maybe even flying in some blimps with equipment...

[Pen-y-gors]

Bean counters again

Shakespeare was wrong, it should have been "The first thing we do, let's kill all the bean counters". They're even more dangerous than the lawyers.

The point about communications for the emergency services is that they work during emergencies. Trying to save a few quid by developing a system that is virtually guaranteed to fail seems a mite silly. Bit like saving on the costs of new police cars by leaving off the wheels.

[Alister]

Re: Bean counters again

Totally agree, this seems to be an ongoing trend for the way government think of all the emergency services.

Fire / Ambulance / Police stations cost too much, lets close all but the bare minimum.

Ambulance / Fire personnel sit around waiting for jobs, that's not efficient, so let's get rid of them.

Hmm, I wonder why response times are going up, can't fathom it out, we've made them more efficient.

[Mage]

Saving x Billion

You can always save money by sacrificing security, resilience, reliability etc.

Using commercial 4G on commercial sSpectrum will certainly save money, but it's not fit for purpose, either day to day or in a disaster scenario.

Handsets and terminals also need a peer to peer mode (walkie talkie mode). You need additional security. Thus the savings will not be as much as claimed. All handsets and vehicle terminals will be x3 to x10 price of regular 4G handsets.

You need low latency push to tall so that "Don't Shoot" doesn't become "Shoot."

It might greatly improve data, but it's a rubbish voice solution. Perhaps data could use 4G and voice the existing system? There wasn't even voice in the original 4G release. It's higher latency than GSM, uses more capacity than 3G and is less robust as it's basically VOIP, rather than a native on air codec.

The only reason at all actually for digital voice codecs on mobile radio is easily encrypt it.

[Commswonk]

Re: Saving x Billion

You need low latency push to tall so that "Don't Shoot" doesn't become "Shoot."

Even with TETRA / Airwave there is no guarantee that the user will be "enabled" when he or she presses the transmit button.

TETRA has a "Grant Tone" which tells the user that they can go ahead and speak; I would be horrified if the EE offering does not provide the same functionality, even allowing for the fact that I hate the word.

The proposed system looks as though it is being specified by politicians and civil servants, not users and engineers who actually understand what they are talking about.

What could possibly go right?

[m0rt]

This is why we can't have nice things.

No one in 'charge' seems to have a clue about provisioning, redundancy, worse cases.

In one sense, I look forward to the brown-outs we are due soon just to give the collective idiots an idea of what they have to consider.

[Anonymous Coward]

"In one sense, I look forward to the brown-outs we are due soon just to give the collective idiots an idea of what they have to consider."

On the other hand, the ones in charge have guaranteed service so won't see the problem themselves and they'll just put it before the taxpayers as a dilemma. What do you want: brownouts, crippling taxes, or fewer essential services?

[Anonymous Coward]

Although this does not improve coverage, there are already steps in place to allow privileged people access to the network when there is an emergency and networks become overloaded. Privileged people have a sim card that is setup with a higher priority. The scheme for evoking the priority is called MTPAS

[Charles 9]

But priorities are all bupkis when it's the TOWER that gets borked. Think "single point of failure". At least most emergency radios can operate independently in a short range: very important in a disaster scenario when there may not be any infrastructure to work with and no time to set up a field station when people are dying all around you.

[Anonymous Coward]

Towers

I can't speak for the police but Fire Service Airwave radios don't have any short range option. There are a limited number of radios on an appliance for use on a fireground but that's it.

The thing about 4G in a built-up area (where the MTPAS system would help) is that there are a lot more base stations in range. Currently you've got maybe one Airwave base stationto a cell and that can cover a largish area.

Having said that the deadline is approaching too quickly. EE have to get planning permission and build/share a lot of towers to get that 97% coverage by 2019.

[NotBob]

Re: Towers

That seems to be a problem, but I don't know that this is a solution. Across the pond we use radios. We have channels which are full duplex for communication with dispatch and the like, but we have basic walkie-talkie channels for tactical and operations channels on scene. Nothing short of an intentional jamming operation is going to take it all down.

[Jamy]

Re: Towers

The Fire Service Airwave Radios are perfectly capable of working in a short range direct operation. Just not used as Airwave charge too much, hence the use of the old analogue Fireground version.

[Commswonk]

Re: Towers

IIRC it's a lot more complicated than that. Fireground communications really have to work. TETRA / Airwave equipment is limited to 1 Watt for "portables" and 3 Watts for vehicle sets, and that simply may not be sufficient for decent building penetration. In TMO (trunk mode) indoor coverage is not guaranteed with Airwave. Why would anyone want to pay a great deal of money for a TETRA radio that might well not work adequately when required in DMO (direct mode) only?

I also have a vague feeling that fire service radios may have to be certified for use in flammable atmospheres, i.e. be Intrinsically Safe, and when I retired (oh shit; it was nearly 10 years ago!) I'm not sure that any Airwave equipment had been designed and manufactured to that standard.

[Simon Rockman]

Re: Towers

There is a lot wrong with the ESN plans but the need for towers is not one of them. Both TETRA and LTE Release 13 have device to device capability. What Release 13 doesn't do is support group calls through a relay. Lots of details here: http://www.theregister.co.uk/2015/01/08/airwave_tetra_switch_off_gov_services_onmishambles/

That article was 18 months ago and the National Audit Office has only just woken up.

Simon.

[Mage]

It's not solved by priority

Mobile towers (2G, 3G, 4G) only work with backhaul links and a complex back end set of computer systems. The "billing" permits the call to be initiated and could even be outsourced to the so called "cloud". If someone forgets to pay a bill or there is an accounting mistake, no calls. Then a second "computer" system manages handing over between masts/sectors (though calls easily dropped then). Specialist voice switches are gateways to other voice systems, edge routers for data.

It's not designed to be reliable or resilient or suit anything other than commercial non-critical calls, other than as a marketing afterthought.

[Anonymous Coward]

It sounds like this will be even better than what the military had to go through with Bowman...

Protip: When you need to spec a comms system that users (and others) life might depend on, make sure you run it by the users...

[m0rt]

"make sure you run it by the users..."

You think GiffGaff should run it?

[ricardian]

It's 13 years since I retired from the Civil Service but for any project (large or small) there was always a Business Assurance Coordinator, a Technical Assurance Coordinator and a User Assurance Coordinator on the project board. The success or failure of a project did depend upon how willing these board members were to stand up and be counted when push came to shove...

[Starace]

It's a stupid idea

Everyone knows it's a stupid idea.

The key requirements are coverage, reliability and voice performance. Instead they've gone for cheap, shiny plus a bit of extra data.

The current solution might be a bit expensive and a bit clunky but it has the key advantage that it actually works unlike this 4G fantasy concept.

The same thinking comes up again and again - why should I spend XXX when my cheap gadget does sort of the same thing? Because maybe your cheap gadget won't work or won't survive?

On a side note I recently had a go with a prototype to try one of the peer to peer forwarding modes for voice in environments without a proper basestation signal. The latency was comical.

[tony2heads]

Re: Radio Amateurs to embed with Emergency services...

Have to agree; if you want resilience get away from digital and if you want coverage go below 100MHz.

This allows anyone to listen, so you may need to use code phrases.

[Commswonk]

Re: Radio Amateurs to embed with Emergency services...

That's the good thing about VHF: From 12-14 miles away, I can easily operate GB3BF (the Bedford VHF repeater) with a small 5W hand-held. It is a one-to-many type approach, however, watching various police shows, that appears to be the case with TETRA.

Not a good comparison. TETRA offers much more than a single wide area channel. And can you work your VHF repeater from every outdoor locality within your quoted 12 - 14 mile radius?

Sorry, but your comment appears to be the case with TETRA does rather show a lack of appreciation of what TETRA does and can do and why, and thus why any replacement system will have to be better than TETRA if it is to be of any use.

Clue: a single TETRA base site with just one carrier can support 3 simultaneous Talkgroups.Or 2 Talkgroups and a single PTP (point to point) call. And so on, including that fact that most sites have more than one carrier...

[Jason Bloomberg]

Costs

The ESN will cost £1.2bn to set up by March 2020 before it starts to realise benefits. But over the next 17 years it is expected to save £3.6bn.

So it''s going to save around £210 million per year. That's loose change in the great scheme of things.

We know that costs of projects rise, potential savings reduce, when a project fails to deliver on time, and we know that's a realistic expectation for any large project.

Is it even worth doing if there aren't good reasons beyond cost saving?

[x 7]

is the EE network going to be hardened against power loss? Here in north Lancashire, the loss of power following the winter floods badly hit the mobile phone networks, which all went offline once the backup batteries flattened. We got about a day of use before they went down, and didn't come back until the emergency generators were put in place.

That was a bloody big emergency, with Lancaster city centre flooded, most roads blocked, no mobile phones, minimal public transport, no power for several days over an extended area, no shops open. Landlines worked, except where people had handsets which used a power supply...........

Each phone mast is going to need a backup power supply that can last can last up to a week. However the fuel and generator is going to be at serious risk of theft from the likes of the TGB caravan dwellers

[SPiT]

There is also the teeny weeny detail that Airwave coverage is defined as in cover if the radio works everywhere and the mobile networks define in cover as the phone works somewhere within 100m. I have seen a specific example of a Policeman doing his job involving extensive use of his Airwave radio in a location where mobile coverage (as used to call them in the first place) required walking 100m up the hill (and away from the incident). It remains to be seen how that will resolve itself.

[Anonymous Coward]

It's a stupid idea

Coverage just isn't up to it anywhere vaguely rural. Even in the relative flatlands of East Anglia, you'll find loads of areas of marginal coverage. Even "villages" of 3000+ people where indoor/groundfloor coverage isn't sufficient to make/not-drop a call. Yes, areas without even reliable 2G voice/SMS coverage on EE.

It's mostly not a case of adding a few more masts, more that there's all sorts of random deadspots because the coverage of one mast doesn't reliably overlap with the next.

And when a cluster of masts in an areas stop working for a week or two at a time (happens every couple of years in my area with EE) then various spots have no outdoor coverage either.

The phone system is incredibly complex with many interconnected systems, and we know full well that large numbers of people across huge geographic areas lose connection for 12 hours or so several times a year, because some server failed.

The basic requirements are fundamentally different.

I think the case made by "x 7" about resilience in the face of longer-term power loss clinches it though. The power needs to keep a dedicated emergency-services mast going, which only operates a few channels intermittently is VASTLY less than a commercial phone-mast (and probably is plausible with a modest battery). No amount of "high priority SIM" helps if the cell-tower batteries have already been run down by Joe Public Facebook 'live' -streaming videos of the flooding or whatever.

For the core, super-resilient service, you want a dedicated system, probably at VHF, to get comprehensive coverage, including indoors, from a modest number of towers. The core system MUST be a 'professional' system and stand apart from "consumer crap".

By all means use commerical networks for non-essential high-bandwidth nice-to-haves (photo/video uploads, face-recognition, whatever) accessories, maybe with custom apps on regular phones.

[x 7]

Re: It's a stupid idea

Just to follow up on the comment re bad coverage on the East Anglian flatlands, the same is true of large areas of the Fylde - the flat drained wetlands between Blackpool, Garstang, Preston and Lancaster. Large areas there have no 2G signal from EE, let alone 3G or 4G. Signal propagation across flat ground would appear to be a significant issue

[David Pearce]

I have worked on a nationwide TETRA network for many years. What costs is the critical network infrastructure - redundant backhaul, standby gensets and fuel logistics, overlapping coverage, indoor coverage.

Commercial 4G will have none of these.

We have mobile base stations for PLANNED deployments and disasters like plane crashes, but they are not a solution for giving coverage NOW to a blue light responding to an emergency.

TETRA terminals are available intrinsically safe for those that can afford them - the price usually scares away most users

All terminals support "Direct Mode" ie walkie talkie style peer to peer and a rebroadcast mode form the handheld users through a higher power vehicle set

[John Robson]

*this*

The cost is vastly in the multiple layers of redundancy built into the Airwave network - that and the network over-provisioning for 'emergency' situations.

On the operational (i.e. radio) side there aren't many opportunities for a single point of failure. Even on the internal side there is a significant amount of resilience (way above any commercial telecoms operator) such that they can actually provide accurate 'call' data to piece together a complex event. Whereas every few weeks it seems some area of a commercial telecoms provider's network disappears for a few days.

Even when the police had to bring in thousands of extra officers (from different regions) to London for the riots - they just selected the right talk group(s) and got on with it. The increase in traffic was substantial, and somewhat extended. I don't see that sort of 'spare' capacity being available from a telecoms operator, they can barely manage to send a text message at or about the New Year.

[David Pearce]

One other elephant in the room is jamming. Cellular jammers are widely available and low power ones are cheap. They are widely used to protect VVIPs from IEDs

TETRA evades jamming by moving the control channel to another carrier and uses an unusual frequency, for which jammers are very hard to find and looking for one will likely make you a "person of interest"

[Anonymous Coward]

Oh? I wonder what would happen if someone set up a powerful pre-Marconi transmitter and made it spark like crazy. Would this be able to jam ALL frequencies indiscriminately?

[David Pearce]

Broadband jammers alias barrage jamming are not very effective as their power is spread over so much bandwidth and therefore diluted.

Targeted jammers go for known control channel frequencies

[Charles 9]

Even if it had A LOT of power behind it or the device has a high gain? Or if it can just sweep frequencies up and down a la a carcinotron? It wouldn't have to jam frequencies all the time, just periodically over short periods. It's the only way you can deal with frequency-agile systems.