– How does the Emergency Alert system work including ‘Cell Level Broadcast’ from Base Station to Handset


– How does the Emergency Alert system work including ‘Cell Level Broadcast’ from Base Station to Handset

In: 7


This is something I know a lot about!

The Emergency Alert System in America works in one of two different ways. The first is via the use of regular broadcast media like TV and radio, and the second is via the use of cellphones, which is much more complicated but surprisingly easier to understand.

Let’s use a tornado warning as an example here. When someone sends out a tornado warning to an area, they will do so using the Emergency Alert System. In the US, every TV and radio station that serves a specific local area is **REQUIRED** by federal law to have a special box at the station to decode messages sent over the Emergency Alert System. This box is so important that if it ever goes down, a station is actually supposed to go off the air within a very short amount of time to correct the problem. In radio specifically, there are two different types of radio stations that have this hardware: Primary entry-point stations, and secondary stations.

A primary entry-point station is one that has a very wide coverage area over a specific market and can be guaranteed to keep running 24/7. This will typically be an AM radio station in most areas. These primary entry-point stations can directly receive information from the emergency alert system via the internet, or by phone lines at any time. When it receives one of these alerts, like a tornado warning, it immediately interrupts whatever is currently being broadcast to play the EAS message.

Now, this tornado warning message has 4 main parts: The SAME header, the attention tone, the action message, and the End-of-Message footer.

The “SAME header” is that weird sounding tone you hear repeated exactly 3 times at the very beginning of an EAS message. This section is actually digitally-encoded data, and it includes information on which physical areas the message is applicable to. The reason it gets repeated 3 times is applicable to how secondary stations receive these messages, and it’s to prevent data loss.

Next is the “Attention Tone”. Basically everyone gets freaked out by this tone, and that’s actually for a very specific reason: This tone was scientifically chosen to sound weird and unnatural to grab the attention of anyone who’s listening. It’s not something you will hear in nature, and this is 110% intended. It’s *very* easy to pick out, and its usage is actually so sacred that this tone cannot be broadcast on radio or TV *at all* unless it’s part of a real EAS message. If you do broadcast this tone outside of an EAS message, there are serious penalties for it. This has been enforced in court several times, too.

Next comes the “Action message”. This is where you’ll hear someone speaking, in plain English, about the exact nature of the alert. This will typically be something like “This is a tornado warning for the following counties: W, X, Y, Z. Seek shelter in a basement, stay away from windows, etc.” This is typically an automated voice, but in more situations nowadays it’s actually just a real voice that someone from a government organization recorded immediately before the alert got sent out. The automated voice is really only used for monthly tests and for alerts that are very common in specific areas.

Finally, you have the “End-of-Message” footer repeated 3 times. This again applies mostly to secondary stations, and it’s once again digitally-encoded data, but it essentially signals to that special box I mentioned before that the EAS message is totally over and the station can return to regular programming automatically.

Now, a secondary station is one that is not a primary entry-point station. Most stations are classified as secondary. This means that they don’t receive EAS messages directly from whoever issued them, but instead they actually receive their messages from other radio stations. Evey secondary station is required by law to listen to at least two other stations in an area at all times for EAS messages, and one of these stations specifically needs to be a local primary entry-point station. (Most radio stations actually listen to 3 stations!) That special box listens for the SAME header from an EAS message, and when it hears one, it will begin decoding that digital data to figure out if the station should broadcast the message or not. Remember, that SAME header includes information about the geographic area that an alert applies to, so if a station in another area hears that message and determines it’s not applicable to itself, it will not broadcast the message.

This process is similar for local TV stations, except these stations will also usually include a visual component so that people who are Deaf can read the text for the alert and still follow it.

Now, cellphones are a bit easier to explain. Cellphone companies are legally required to offer a special place for local governments and organizations to send out EAS messages. The message that gets sent to the cellphone company includes basically all the same data that a message sent over radio or TV would include. The cellphone company then processes this data in much the same way as a TV or radio station. Since it also includes location information, the cellphone company will then blast that alert out to all cellphone towers in the affected area. The actual cellphone tower then sends a message to all cellphones currently connected to it that includes the content of the EAS message. The cellphones connected to the tower then display this message!

It sounds complicated, but it’s actually really amazing that we have something like this, and that it works so well and consistently! The next time you hear an EAS message on the radio or TV, listen for those 4 distinct parts and you’ll definitely be able to pick them out!

Every phone on a cell network is always listening to a paging channel. The main CPU may sleep for a long time but the modem wakes up every second, decodes the paging channel, and goes to sleep in 10-20 milliseconds if nothing relevant to it was broadcast.

That is how incoming calls and push notifications work. Cell tower broadcasts on the paging channel phone ids which should wake up and check for incoming calls and data. To wake up all phones cell tower broadcasts on the paging channel special id which mean every phone should wake up and decode special cell broadcast data in the data channel. To prevent overwhelming the tower unlike incoming calls and push notifications cell broadcast data does not require each cell phone to request any data from the network. All phones operate in receive-only mode. If a phone fails to receive the broadcast it cannot request it. Cell towers can be configured to repeat the same broadcast several times so that phones have multiple opportunities to receive the data. Each message has an id so the phones that already received it once successfully can ignore retransmissions.

Authorities that wish to broadcast a message coordinate the list of areas with the carriers ahead of time. For example a US state may provide a list of counties and a list of major cities with canonical spelling. The carriers compile a list of towers in each area. When it is time to broadcast a message across a list of areas carrier merges the lists of towers in the requested areas and tells each tower to broadcast the message.

In many cases especially in rural areas cell tower coverage does not match the true boundaries of areas. About 7-10 years ago to make notifications match the true boundaries the FCC requested an enhancement. Now a broadcast message can have an area defined as a polygon (list of latitude, longitude points). Once phone decodes the broadcast it checks if it is actually in the area. If it is not it goes to sleep.