Radio traffic statistics
Given that my scanner project has been running for about seven months, it seems like it's ripe for analysis. I started wondering just how busy the city's radio system is, both in terms of who's talking (police, fire, electric, and so on) and which frequencies they're using to talk.
The top 20 list of talk groups by call count looks like this:
|75911||PD 4 - Records||112|
|5440||Parks and Rec||880|
|4874||Electric 5 - Maintenance||2032|
|2213||Fire 5 - Training/Drills||1776|
|991||Water treatment 2||2096|
|928||Comm Techs Services||432|
|843||Event 1 - Art and Wine Festival||16016|
|513||Public Works - Automotive||752|
|432||PD Special Ops 1||1072|
In terms of frequencies, there are only seven. Some of them are used as control channels, which means they blast a stream of data continuously for about 24 hours. Then it rotates to another frequency and the old one is now clear to handle voice calls.
As a result, those frequencies which never host the control channel are always eligible to have a voice call assigned and show up more often. There may be a little unbalancing here due to maintenance or other factors which take one or more transmitters down for an extended period. There was a time when every call on a specific transmitter had horrible audio quality one day, and it vanished from the list for a while.
There are other things you can learn by running queries against this table. The busiest hour of the day by call volume on the main police channel (PD 1) is from noon to 1 PM. The least busy hour is from 5 to 6 AM. Their busiest days are Friday and Saturday, with the rest of the week being relatively calm in comparison.
They have a special talkgroup which goes out to all units simultaneously called "PD Wide". This is used to find a unit who's not answering calls on the normal channel, to restrict traffic or call for assistance. Most calls on that particular talkgroup happen between either 2 and 3 PM or 10 and 11 PM. Its quiet times are between 6 and 8 AM. Strangely, in terms of days, it's heaviest on Thursdays and lightest on Wednesdays.
I'm tempted to connect this to something which would build tables like my old support ticket inflow stuff just to see what else pops out of the data. There might be some surprising trends just waiting to be found.