BCSO Radio System Overload during Parkland Shooting 2/14/2018

[RFI-EMI-GUY]

Very interesting.

https://soundcloud.com/user-150674788/broward-county-radio-problems-feb-14-2018

 

[AlphaFive]

BCSO

That's cute.. They used to simply say "Motorola has been paged" and leave it at that. The dreaded word CRASHED was never used in Seminole County,, nice find.

 

[mmckenna]

A couple of things:

An officer keying up the system to say he can't key up the system. As they say, if you are not part of the solution, you are part of the problem.

Officers need more than 45 seconds of training on how to properly use their radios. Until agencies figure this out, there will keep being issues.
And even when they do, there will keep being issues. Part of that training needs to be an explanation that this is going to happen, and how to handle it. Tying up a talk group to complain about the system being overloaded is not helping.

 

[RFI-EMI-GUY]

A huge problem with digital systems is that there is a great amount of latency. It is easily dismissed by the vendors as meeting P25 standards, but it really adds up over a shift. In every exchange between dispatch and a subscriber their is wasted air time. In fact P25 FDMA has negative spectral efficiency when you factor this in. All that dead time slows down traffic. I am not sure P25 even addresses latency, rather access time.

If you have a hard time wrapping your head around this, watch a newscaster talking with a field reporter over a digital studio to remote audio path. There is always that painful dead air after the anchor person says "over to Bill at the crash scene for more". Bill always seems kind of lost for a second or two.

Then there is the problem of "doubling". Two digital subscriber units should never get granted access to the channel at the same time. To prevent this there is an arbitration timer. Someone always gets "bonked" when the talkgroup has a lot of traffic.

Now, that being said, the control channel has to process a ton of requests in an emergency. Multiple calls, new TG affiliations, etc, . To the extent BCSO system actually "crashed" under that demand is unknown. One would think that Motorola has stress tested the control channel to deal with a high volume of requests. Maybe a problem at the IP transport level? Some underlying design or configuration error?

A commercial consultant over at APCO's user group, opined that they should be switching traffic to conventional mutual aid channels during a crisis like this. I don't share that opinion. That would only create more havoc for responders and dispatcher. If that is the solution, then P25 has been the wrong technology for public safety.




Sent from my SM-T350 using Tapatalk

 

[mmckenna]

A commercial consultant over at APCO's user group, opined that they should be switching traffic to conventional mutual aid channels during a crisis like this. I don't share that opinion. That would only create more havoc for responders and dispatcher.

I read his post the other night, and I agree with you. I believe that would have caused a lot more issues.

If that is the solution, then P25 has been the wrong technology for public safety.

It is. I remember an APCO guy getting rather upset with me when we decided to not use the P25 standard for our systems. We kept our PD and Fire analog on a conventional system, and split the non-public safety users over to a non-P25 trunked system. APCO has a lot of time, money and effort into P25 and they tend to defend it pretty hard. Between them and the manufacturers, it's not doing anyone any favors. The idea was that prices on the gear would drop quickly when it was adopted widely. It hasn't.

 

[RFI-EMI-GUY]

I read his post the other night, and I agree with you. I believe that would have caused a lot more issues.



It is. I remember an APCO guy getting rather upset with me when we decided to not use the P25 standard for our systems. We kept our PD and Fire analog on a conventional system, and split the non-public safety users over to a non-P25 trunked system. APCO has a lot of time, money and effort into P25 and they tend to defend it pretty hard. Between them and the manufacturers, it's not doing anyone any favors. The idea was that prices on the gear would drop quickly when it was adopted widely. It hasn't.

Sadly, even if you factor in the highest percentage matching federal grants for purchasing P25, they never came close to reducing the cost to the level of analog. There is no cost savings and I would argue, really no technology advantages. DMR has the lead in cost benefit ratio at this point in time. The VENDORS and APCO are complicit with distorting and twisting reality to fit their vision of P25.

This all said, I went back and listened to the tape again, and they are still operating analog Smartnet/zone at BCSO. The problem with units doubling and potentially some control channel saturation issues remain, but apart from poor maintenance (Motorola has the system at EOL), no reason for the system to "crash" under load. Imagine how much worse it would be with P25 digital and the latency delays and bonks.

I heard/read a rumor that they may be in process of a P25 upgrade so there could be analog channels out of service or Smart X switch problems.

 

[mmckenna]

and I see your post on there too, Joe.

No amount of technology is going to fix it. Single talk path, only so many users. Unfortunately there isn't any money to be made in training officers in properly using their radios. It's so much more lucrative to sell new technology.

 

[Bote]

How does a P25 public safety system handle a second transmission on a talkgroup that is currently occupied?

Does it deny the guy trying to step on the transmission or does the second radio get a channel grant and let God sort 'em out?

What about simultaneous key-ups?

 

[RFI-EMI-GUY]

How does a P25 public safety system handle a second transmission on a talkgroup that is currently occupied?

Does it deny the guy trying to step on the transmission or does the second radio get a channel grant and let God sort 'em out?

What about simultaneous key-ups?

The P25 system does not allow a second radio to transmit . The second user will get a short bonk, and his radio will unmute and hear the first transmission in progress. If the dispatcher is talking, a first subscriber radio may talk, but his audio will be heard only by the dispatcher, dispatchers have "wireline priority".

In an analog system, the same feature is available to prevent doubling, however if I recall it is an optional setting and does not fully preclude doubling while a conversation is already taking place.

The BCSO system is still operating 3600 analog. An upgrade is in process.

I got some feedback on the BCSO problem today and it is being considered that there were so many affiliations from other responding agencies switching channels that the inbound control channel was briefly overloaded.

The inbound signalling word from a subscriber is sent up to 16 random times if there is a collision on the inbound control channel (slotted ALOHA), so there can be a cascading effect from multiple affiliations and PTT requests.

Motorola had used a word describing "System Throttling", but I don't know if that is to describe the effect of these collisions, or there is some active feedback mechanism to deal with it. Time to look at Motorola patents.



Sent from my SM-T350 using Tapatalk