Boston Officials Did Not Shut Down Cell Network After Marathon Bombing

An anonymous reader writes with this excerpt from Motherboard about the immediate aftermath of yesterday's bomb attack in Boston, which attempts to explain the (unsurprisingly) poor accessibility of the cellular network after the blasts: "Gut instinct suggests that the network must've been overloaded with people trying to find loved ones. At first, the Associated Press said it was a concerted effort to prevent any remote detonators from being used, citing a law enforcement official. After some disputed that report, the AP reversed its report, citing officials from Verizon and Sprint who said they'd never had a request to shut down the network, and who blamed slowdowns on heavy load. (Motherboard's Derek Mead was able to send text messages to both his sister and her boyfriend, who were very near the finish line, shortly after the bombing, which suggests that networks were never totally shut down. Still, shutting down cell phone networks to prevent remote detonation wouldn't be without precedent: It is a common tactic in Pakistan, where bombings happen with regularity.)"

Comments

That doesn't mean it wasnt jammed

[headhot]

Why the network operators didn't get requests to shutdown the network, that doesn't mean it wasn't jammed. The military has jammers it uses where they suspect IEDs to prevent triggering via the cell network. There is no reason why the BPD, DHS or other agency would not have jammers for such an occasion. I would be surprised if they did not with all the money that was thrown around after 9/11

Re:That doesn't mean it wasnt jammed

[usuallylost]

I am leaning toward the 50,000 people in the area all trying to call home at once.

Re:That doesn't mean it wasnt jammed

[MightyYar]

I was able to text back and forth with my niece, who was at the race. No idea what network she is on, though the interwebs say it is an AT&T number. I don't think things were being actively jammed.

Re:That doesn't mean it wasnt jammed

[tedgyz]

Right. We had similar problems in DC during the Rally for Sanity.

Re:That doesn't mean it wasnt jammed

[TheCarp]

> There is no reason why the BPD, DHS or other agency would not have jammers for such an occasion.

Really? And why should they? The entire idea that they should have them is based on specific technical details of specific attacks, and requires both that they guess right that its the right time to use them and that the bomb maker didn't anticipate their use.

Additionally, with all the people involved, they generally want people to get the "Im safe" messages out, because it decreases overall mayhem and people trying to contact them for information.

> I would be surprised if they did not with all the money that was thrown around after 9/11

Well I wouldn't either, but, thats a different issue.

Re:That doesn't mean it wasnt jammed

[MickyTheIdiot]

...or they aren't engineered for the once-in-two-decade major catastrophe. Technology has limits, and you have to draw a line somewhere so that people can actually AFFORD to use the service.

Re:That doesn't mean it wasnt jammed

[Lumpy]

AT&T networks crumble when there is only 10,000 people in a small area like that. WE have a small motorcycle rally here in July and the 8000 to 10,000 people will utterly crush the nearest 3 cell sites for the entire weekend for AT&T. Verizon does better, but data is completely useless.

Re:That doesn't mean it wasnt jammed

[Gilmoure]

SMS is one of the lower level cell phone protocols, uses the least bit of bandwidth and is almost always on, even when higher level voice and data fail. Ideally, modern phones could be set up to pass SMS traffic from phone to phone, when a tower signal is unavailable. This would allow messages to get in and out of disaster areas like New Orleans during Katrina.

Re:That doesn't mean it wasnt jammed

[sycodon]

That's pretty much the definition of "jammed".

Re:That doesn't mean it wasnt jammed

[MightyYar]

Presumably, any jamming worth a crap would need to block SMS, which is why I mentioned it.

Re:That doesn't mean it wasnt jammed

[Anonymous Coward]

when I worked in telecom, specing out a CO for site we would generally have enough T1's (24 full-duplex lines) to provide for 20% of the population as that would cover normal traffic at any one point in time. It was always known that during emergencies it would become overwhelmed and low priority calls (you and me) would fail while there are settings to allow high priority calls (emergency responders, police, government, etc) to drop a line in use by someone lower priority and go through.

My bet, is the majority of people starting phoning home to let loved ones know they were all right - the networks aren't designed for that level of instantaneous traffic.

Re: That doesn't mean it wasnt jammed

[RivenAleem]

I could be accused of stating the obvious. It could be set to turn on all the time and we'd be done with carriers.

Re:That doesn't mean it wasnt jammed

[Jane Q. Public]

"... and you have to draw a line somewhere so that people can actually AFFORD to use the service."

That's easy: give CEOs realistic paychecks.

not all that effective

[O('_')O_Bush]

I would think that shutting down cell towers wouldn't be particularly effective, given that the same mechanism that would allow one to trigger a bomb with a cell phone is also present in other RF devices such as baby monitors and walkie-talkies.

Re:not all that effective

[dunezone]

But baby monitors and walkie-talkies can pick up interference and other communication which can cause the trigger to go prematurely. With a cell phone its waiting to receive a phone call so if no one knows the number its less likely to go off prematurely.

Re:not all that effective

[Anonymous Coward]

Telemarketing to cellphone numbers is now an anti-terrorism tool.

Re:not all that effective

[geekoid]

Yes, it's the phone. Not the person not paying attention~

Re:not all that effective

[egamma]

I would think that shutting down cell towers wouldn't be particularly effective, given that the same mechanism that would allow one to trigger a bomb with a cell phone is also present in other RF devices such as baby monitors and walkie-talkies.

They could also set the bomb to trigger if it loses reception.

Re:not all that effective

[TheCarp]

Totally different threat profile.

A convoy is moving and is a very small target in a very large area. It is especially exposed, and an especially juicy target in a war zone. You can expect attacks fairly frequently, they have to find you/be ready for you.

This event is predefined, the attacker knows where and when the targets will be there. The attacker already has time to prepare and makes himself known on his time table.

This changes everything. In your convoy for example, there is no benefit to rigging bombs to blow when their signal is jammed or even to arm in response to signals from a jammer.... as the prowler is not the convoy and need not be all that close to them, arming or blowing in response to the jammer means wasted bombs or blowing up innocent bystanders, will almost never hit a convoy.

Here we have a totally different scenario. A secondary device triggered by a loss of signal could have huge impact. The devices are already at their pre-determined target. You don't jam, he can detonate, you do jam, they might detonate, point is....you have no way of ever knowing what he planned until its all over.

Overloaded

[Anonymous Coward]

After the Earthquake in Virginia in 2011, you couldn't make a cell call to save your life, since several million people picked up their phones within a few minutes of each other. Text messages went through fine within a half a minute or so. Something similar happens whenever an unexpected event of note happens anywhere.

Re:Overloaded

[rvw]

And I don't think there is an engineering solution for it. It's a race condition... there will always be a bigger event that needs more capacity and you end up with a huge, costly network no one can afford to use and, even if everyone could afford it, would be have massive capacity.

I think you wall off some capacity for emergency users (911, police, first responders) and do your best with the rest.

One solution would be to allow text/sms only. The phone keeps trying till it has delivered the message, it's small in size. You could even send all users a broadcast sms to let them know that.

Re:Overloaded

[tlhIngan]

One solution would be to allow text/sms only. The phone keeps trying till it has delivered the message, it's small in size. You could even send all users a broadcast sms to let them know that.

You do realize that control channel overload is what causes the cell network to go down these days, right?

When the control channel is overloaded, a phone can't make a voice of data call (requires using the control channel to select the appropriate voice or data channel and timeslot).

Allowing text/sms only would work, but it also means that the rest of the cell tower is underutilized. Better would be to reprioritize things so re-establishing a data channel, voice channel, or handoffs have higher priorities than texts to get better utilization. Texts can always be delayed and the phone retries anyways, while being able to establish a voice and data connection can come in handy.

And yes, networks often have 911 prioritization as well - establishing an emergency call can use reserved bandwidth so even the fullest of cell towers can handle an emergency despite being slammed with traffic.

It's why the iPhone basically killed AT&T's network - the Infineon chipset it used was very power efficient and was aggressive - the instant no data was being sent, it tore down the data connection. Doing this often enough consumed control channel traffic until AT&T's network was overwhelmed with administrative data. It lead to one of the oddest conclusions available - AT&T had the fastest data network around. IF you could establish the connection. The voice and data channels were underutilized, while the control channels were overloaded so you couldn't make a voice or data connection, do handoffs (leading to dropped calls), and texts got delayed.

A poorly-written Android IM app on T-mo in the early days did the same as well.

Though, you'd think prior to a handoff, when the phone is scanning for a new tower, you could have the phone use the old tower's control channel to communicate with the new tower to get new voice/data channel allocations for handoff while the phone tries to re-establish the control channel connection. Add some in-band signalling just in case and you could offload some control channel traffic.

That is the best use of text messaging

[damn_registrars]

I almost never use text messaging, but it is extremely useful when cell networks are overloaded as it uses almost no bandwidth and hence messages almost always get through. Unfortunately not many people think of it that way and tend to keep trying to make a voice call when the network is rejecting their attempts.

Re:That is the best use of text messaging

[afidel]

SMS actually take zero bandwidth on GSM networks, they use the ping packets that the phone must exchange with the tower every so often to send the message, it would otherwise be padded with zero's. That's why the message length on SMS is so short, it's limited by the difference between the header needed for a ping and the size of a timeslice. Some phones will opt to use the data network if available to ensure faster delivery but SMS was really a brilliant hack to take advantage of the nature of the network.

it should be common knowledge

[nimbius]

that during an immediate crisis or disaster, cellular networks will become quickly overloaded. network providers acknowledge this and tout things like COW and COLT (Cellular on Wheels and Cellular on Light Truck) as a solution. what isnt clearly stated is that these systems may be hundreds of miles from the immediate area, or may rely on existing trunks and uplinks that are themselves completely saturated, if they havent been destroyed by $crisis || $disaster. cellular providers also have a terrible habit of booking these emergency systems for sporting events to augment their second rate cellular networks.

For geeks who understand how cellular works its limitations are pretty obvious, so im seriously wondering if amateur radio played any part in assisting during this crisis?
ham was designed from its inception to help in a civil emergency, and it would be hard to imagine an event like the boston marathon without at least one 2 meter or 6 meter general class or extra present.

does boston use ASTRO? or EDACS radio networks for emergency services? if so how did these networks perform?

Re:it should be common knowledge

[cdwiegand]

They weren't activated, and the ones that were in use during the race (for coordination) were all evacuated and followed those orders (http://cqnewsroom.blogspot.com/2013/04/boston-marathon-update-all-hams.html)

You're more likely to find hams helping in inter-departmental capacity, where large-scale (this was so not large scale) events require coordination between multiple police and fire departments, hospitals, etc. This was a local situation where Boston Police (and to a point DHS) were involved, but no other agencies - they can usually handle talking on their own radios to themselves.

Re:Text Messages

[Anonymous Coward]

That technique seemed to work for my last relationship.

Dumb thing to focus on.

[Anonymous Coward]

Whether cell towers were working or not is a stupid thing to focus on, here. How about the coming absolute surrender of all remaining liberties? Since 9-11, I've repeatedly pointed out that all we need is one more big terrorist event to shake the population enough that we will give up everything. Complaits about the TSA, second amendment, privacy, government and corporate wiretapping without justification. All of it. It is over. We lose.by attacking us, we shell up. We take away our own freedomFOR them. It is time to shutter yro.slashdot, because it no longer matters.

Re:Could be cell phone

[Nidi62]

My first thought was that they were simple countdown timers, as there were a couple seconds between explosions. If they were triggered remotely, they would have gone off simultaneously of they were triggered by the same operation,or longer than a few seconds between explosions if they were triggered separately (I guess the bomber could have had 2 phones and dialed the bombs as the same time, but that seems overly complicated). The amount of time between explosions lends credence to the idea that they were simple timers: set one then set the other a couple hours before, drop them off, then get away.

Re:Could be cell phone

[geekoid]

By 'so many' you mean 'none'? HINT: don't pay attention to the news to get any good information about the bombs for at least 24 hours. There reporting anything they hear, regardless of source.

Then upgrade the cell network

[HeckRuler]

So... considering that's we hear about this with EVERY major catastrophe, would this be the sort of national infrastructure concern that we would want to mandate that the cell companies install extra capacity? You know, in case of emergencies. Are we at the point that we can consider cellular connection, or generically wireless connection, to be a basic utility and not a cutting edge hip new ordeal that only the rich can afford?

And hey, since they've got ALL THAT BANDWIDTH, just lying about in case shit hits the fan, it'd be great to sell it on the cheap. You know, that idea that society and the fundamental utilities is here to foster growth rather than wringing out the last coin from the customer's pockets.

Re:Then upgrade the cell network

[HeckRuler]

Utilities, scarcity, and the rates they charge. Remember Enron? Remember how they were given control of power generation? The idea was that with the magic of the free market and competition that prices would come down. It turns out that if there's a fundamental need for a resource, like a utility, the people that control said resource strive to keep it scarce so they can justify exorbitant prices for it. Now, that might all be different if we have a large host of people competing to eat each other's lunch. But thank god we've got a lot of wireless carriers to choose from, so THAT'LL never happen.

If bandwidth is scarce, they'll charge more. If bandwidth is plentiful, if we have plenty of supply, they'll charge less. And it's not that some exec will see that they have excess supply and start a sale, no. You'll have some accountant see that they could take all the customers in district X and still be functional, even if they sold at a lower price than normal. All they have to do is undercut the competition. The only difference to them is extra paying customers at the end of the day. They have the infrastructure already in place.

But no, this is not without side effect. There's an initial big expense as the phone companies upgrade their infrastructure. I'm saying the cost of installing extra bandwidth is negligible compared to the benefit of getting the wireless carriers to compete with each other and having a utility that actually functions when we need it most.

Short memories

[MickyTheIdiot]

Uh.. doesn't this happen after just about every disaster?

If you design the networks to work at the utilization that you see after a disaster there would be cell phone towers at every corner, our bills would be $500 or more a month, and it would be using a very low percentage of its capacity 99.99% of the time.

It isn't what is important at the moment, anyway.

SMS uses a different protocol

[sl4shd0rk]

I believe SMS piggy-backs on a transmission from the tower which is a different protocol than what is used for voice/data*. It seems possible that SMS may work when voice/data has been blocked.

"transport messages on the signaling paths needed to control the telephone traffic during time periods when no signaling traffic existed. "

[*] - http://en.wikipedia.org/wiki/Short_Message_Service

Re:SMS uses a different protocol

[isorox]

Which is why charges of ANYTHING for text/sms should be fraud! Cellular would not work without this data stream, they act as if they are providing you with a service.

Don't like it? Don't use it.

Re:Could be cell phone

[wonkey_monkey]

Officials also announced a twist in the probe: Suspicious packages that were detonated out of precaution were not explosive devices after all.

That's not a twist, it's just a thing. A twist is if it turns out to have been Richard Simmons.

VZW appeared overloaded, not blocked

[CLorox]

VZW appeared heavily overloaded and calls were not going through. Additionally, text messages also appeared to be throttled or heavily delayed. If this was a result of jamming or some other technology to throttle the network, calls were being placed, they were not however providing audio. I received about 20 calls from my girlfriend who lives in the area and her calls were ringing through and "completing", but no audio was making it over the line. Calls I was placing appeared to ring through (five or six rings?) and made it to voicemail in most cases, although I did get a couple Verizon messages instead of the voicemail box.

Text messages we were sending each other were either extremely delayed or never made it at all (some did). I would go with the disaster norm of badly overloaded. We resorted to email via wifi instead of relying on the cell networks. When she took to the car to pick up her sister in South Boston (T services were shutdown in and around Boston), she was able to start completing calls and texts were making it through.

Re:Something's weird here

[kannibal_klown]

As someone who had a loved one in the World Trade Center on 9/11 and couldn't reach her on her mobile until the afternoon, I'm not surprised that a cell network became overloaded. It's happened in other times too.

Considering this wreaked of terrorism, especially to those on the scene, things really blew up. Marathoners calling loved ones and even just REGULAR citizens in Boston (perhaps far away from the site) getting / making calls in a panic, etc.

Cell towers aren't magic, they can only support so much. And since the phone companies aren't using their profits to expand their existing networks then a large metro getting hit with an event is going to overload it... or at least a general region. I mean, in that immediate area alone you had: people living / working in the buildings, LOTS of runners, LOTS of spectators, etc.

Perhaps the feds DID put up a jammer. Perhaps we don't know the whole story. I'm just saying, that an overloaded network sounds perfectly plausible.

Re:Could be cell phone

[rvw]

Depends. There are reason to delay explosions.

That is true, but if you are going to delay explosions, you don't delay them for only a couple seconds.

If you want the simultaneous, then it doesn't matter if they are a couple of seconds apart. For effectiveness it has no consequenses. And maybe this is easier to handle, e.g. if you have to use two remote detonators. If bomb A doesn't go off, you can decide not to use bomb B, for whatever reason.

Re:Something's weird here

[rvw]

Something's really weird going on with this whole mess. Given redundancies and failover capacity, I'm having a hard time believing that simple load caused failures that blocked cellular transmissions, especially as the failure occurred pretty much right after the blast, and not enough awareness would have been out there to cause the level of traffic needed to bring down the cell system soon enough to have had any effects in blocking cellular-detonated explosives.

Given the fact that a simple new year can bring down a network (in the sense that most people cannot make calls for half an hour), this doesn't surprise me.

Re:They didn't? Why not?

[geekoid]

Becasue it was over.
Plus, any response that is standard will be trivially worked around.

Dated information

[Brucelet]

I don't know for sure on cell service, but there's a whole lot of other outdated information in that article. Specifically, the fire at the jfk library is now known to be unrelated, and law enforcement officials have stated that no undetonated devices were found.

Media

[isorox]

* Media spouts a load of crap to get ratings
* News at 11

Oh wait, not news at 11, it's news all the time. Doesn't matter that there's nothing new to report, that it's all over, that it will takes days to get any more answer, we have to have wallpaper news because all the other stations are having it. Everyone's hoping for another 9/11.

Oh how I wish for a return to a half hour news bulletin 3 times a day, when journalists had time to go out and find what's going on rather than sit in a studio doing two-ways, reading wires and copy that's come from the studio.

That's enough from me standing outside an empty office block, back to you in the studio. I'll be here again in 15 minutes though for an update.

Re:Ah, the rush to misreport

[isorox]

Ah, the rush to misreport. This is why I wait a week to read about what happened. News outlets, and law enforcement officials, will just make stuff up because it's too early to know what really happened. Enjoy your speculation!

Facts: There were 2 explosions near the finish of the Boston Marathon yesterday. 3 people have died.
Opinion: Everything else on the news for the last 24 hours.

Use SMS in emergencies

[mrops]

In such emergencies, its better to use SMS than place a voice call.

SMS rides on control signal and as long as your cell phone has a signal, it will get queued and delivered.

Voice calls require acquiring of a dedicated voice channel, these are limited and overloaded in such emergencies.

Re:Use SMS in emergencies

[crakbone]

After Katrina I had a friend who was worried about their father and could not get through the cell network. I recommended SMS she got through but all he said was call me.

"Use SMS in emergencies" is actually correct

[SpaceManFlip]

Awhile back, you would see disclaimers here and there saying "the reliability of SMS messaging should not be counted on in emergency situations" or something to that effect.

But the opposite actually appears to be true, when the "emergency" is in an area where a lot of folks are trying to use their cell phones at the same time.
During some weather catastrophes a few years back, I could not get a single call to go through on the biggest carrier in my area, but texts did go through successfully. It's just less data packets to carry the information.

It's also interesting to note that the failures of cell networks illustrates the cell carriers overselling of their networks, or to say it a different way, their failure to invest in their infrastructure to adequately support the number of customers they have.