A $47B Emergency-Response Network That’s Already Obsolete | Hacker News Reader

A $47B Emergency-Response Network That’s Already Obsolete

theatlantic.com

102 pointsprostoalex23 comments

This article is from 2016. AT&T already won and California was last state to opt in: http://urgentcomm.com/ntiafirstnet/it-s-unanimous-california...

Thanks, this has been up for 8 hours and it still doesn't have (2016) after it

I think the supposition of this article is that dedicated push-to-talk infrastructure is obsolete - something I feel is demonstrably false.

PTT only networks are much easier, and less infrastructure heavy to keep operational, they also have significantly lower backhaul requirements. In addition, they have better overall reliability during disaster situations.

Motorola (alone, they are one of 5 big players in the industry) ships something like 500m+ in two way radio infrastructure yearly, and its not really a shrinking market.

So yes, FirstNet is a boondoggle, and one of those 9/11 projects that should die or go a different direction, to a great extent the interoperability they offer is fantastic, if anyone could ever use it - but two way radio? not dead, and still very relevant for certain industries and services.

Just curious, why are ptt networks less infrastructure heavy?

This question is a bit out my domain, but I believe the answer boils down to being able to re-use one antenna for send and receive with push-to-talk, but for full-duplex communication you either need another antenna, or more complex signal processing.

Strictly speaking you can use half as much bandwidth by having TX and RX share it, i.e. half-duplex versus full-duplex. It can also somewhat simplify transciever designs in a number of ways. It is also a fairly straightforward model: if you're on the channel and in range of the transmitter or a relay (and maybe have the right preshared key), then you will hear and be able to transmit on that channel.

The FirstNet is/should be unrelated, as far as I'm aware. I think it's supposed to be for data applications like image and document sharing.

a bunch of reasons.

I'm working on a state wide radio system for a power company, its a simulcast network, with 20 talkpaths. Your worse case bandwidth utilization is on the order of roughly 1.5 megabit a second per site (realistically, its much lower). In addition, if a site fails the other ones will still function, also, if the network core fails, the sites go into a failsoft mode, which allows single site communications to still occur.

FirstNet is based on LTE. Last time i looked into it, it was a private LTE network, with the ability to piggy back on public LTE when there's no private coverage (with a high priority). So you can forget about the simplicity of a basic PTT network. Here, it's MCPTT (Mission Critical Push-To-Talk), which is a layer on top of LTE leveraging the LTE multicast support (eMBMS) when applicable. So not exactly super simple.

There a protocol extensions for public safety on top of MCPTT. To better handle out of coverage situation (D2D or ProSe, which is device-to-device communication with relay support too), to hook custom applications on the network side, etc. But the infra is LTE, so at least it leverages an existing technology. Still, with some additions on top to better support public safety use cases.

And the "obsolete" part is not very convincing. Yes, there's a lot of hype about 5G but LTE will be long lived and perfectly fine for such an application for a long time. I can't comment on the overall efficiency of the FirstNet program, but the idea of leveraging LTE instead of rolling a custom tech or keeping using old and inefficient ones seems sound to me.

Hell, 3G and EDGE/1x are plenty capable today for a lot of use cases

> ptt networks less infrastructure heavy

That's often walkie talkies, no? Those are pretty handy in first responder situations.

Nation wide networks used in Netherlands: https://en.wikipedia.org/wiki/P2000_(network) https://en.wikipedia.org/wiki/Terrestrial_Trunked_Radio#cite...

P2000 is basically to receive a text. The C2000/TTR is encrypted, meant for Policy/Ambulance/Fire departments and apparently used by many countries in the world. It'll work even when other networks are down. Apparently fully used since 2007.

I'm a first responder for my building, it used to be that the C2000 system (simply stated: "professional walkie walkies") would be difficult to use in my building (couldn't go through concrete). We'd lend them our walkie talkies (they work much better as we have a support transmitter within the building). Over the last few years they seem to have improved C2000, nowadays they don't need ours for most areas.

The various problems with the C2000 system are explained in Dutch (!) at https://nl.wikipedia.org/wiki/C2000#Problemen.

English summary:

- Connectivity can be lost when entering a building (e.g. by firemen). For this reason, direct analogue radio connections are often used.

- The system can't handle the load during large calamities.

The ministry of home affairs is investigating.

> The system can't handle the load during large calamities.

For a system that's supposed to be used during one I'd say that's a rather large problem.

Compared to cellular, ptt networks are often vhf frequencies (<200 MHz) and higher power (>5 Watts), therefore coverage from a mountain top repeater is practically infinite compared to cellular, and limited only by terrain, requiring far fewer base stations and repeaters and backhaul for the same coverage.

Agreed. Half the value IMO is in the UI. A smartphone is an awful user experience with gloved hands, and demands your attention while a good radio mostly gets out of the way. Radios using a LTE-based dedicated network like they mentioned might be fine, but at that point we've really just changed the operating frequency.

How does something like this actually cost $50B?

Is there any transparency/breakdown that's available? I'm just confused as to what the money was actually spent on...

> According to the GAO, estimates of its cost range from $12 billion to $47 billion

I understand the desire to really drill home the waste involved, however when the cost range is $12 to $47 billion and the article chooses to put $47 billion in the headline, it's at the edge of clickbait.

There is the definite possibility that it actually costs something like $69 billion, in fairness.

That issue (communication between multiple first-responder branches) probably isn't USA-specific. How is it solved in other countries?

https://en.wikipedia.org/wiki/Terrestrial_Trunked_Radio

It's getting outdated though. In Sweden there's talk about a nation-wide LTE network specifically for first responders. Cheaper and better, I presume.

Why not let the US Digital Services be asked to step in for this project? They could either save the project or abandon what has been done and start afresh with a better and leaner approach. Or they could make an assessment whether the solution being envisioned is still relevant to the current situation.

I'd worry about their ability to do field deployments, this isnt a datacenter deployment, its 50,000 cell sites nation wide.

some how I think the US DS is not resourced to do this sort of project it's not a simple website. All they could do is assuming they have enough senior resource from telecoms as project manage some one else doing it.

They were not resourced initially when they had the idea of tackling some of the government's most expensive and daunting challenges. Maybe what I am saying here is that, given their solid objective and passion of making a difference and turning things around, they could do the same to this project as well. They have processes in place which could help with selecting people and technology that could best resolve the job at hand and at the same time, cost less whilst maintaining a high degree of quality.

Reply on news.ycombinator.com