Understanding Capacity Plus trunking

[IanWraith]

Hello
In answer to indigo88's post the bit 2 in the CSBKO=62 PDUs refers to the channel in use. So the CSBKO=62 may be being sent on channel 1 but bit 2 tell the radios which channel the voice or data activity is taking place in channel 2. I'm finding the channel number from the CACH.

In addition I can add that bit 7 is only set then nothing is happening. When there is group , unit to unit voice or data going on then bit 0 is clear.

Also indigo88's mention of hearing what sounds like a control channel i.e short bursts of data on a single channel is similar to what I have heard occurs on a Capacity Plus system in the south of England. However the 4 channel Capacity Plus system near here doesn't have such a channel. Perhaps because of that on this system I only see CSBKO=62 PDUs when there is activity on that particular frequency. So how on earth does a radio sat on say channel 5 know there is a group call that should involve it on channel 7 ? Do they scan all available channels when not involved in a call ?

I have a nasty feeling their are many types of Capacity Plus systems.

Something else that appears to me to be missing from Capacity Plus is any kind of a system ID. Both MPT1327 and the more modern Tetra standards transmit what is supposed to be a unique system identifier so that radios know if they should even try to connect to a system.

Oh and in the local system I mentioned bits 11 through 15 are always zero. Confused - I am !

Regards

Ian

 

[inigo88]

Hi Ian!

I spent the better part of a night pouring through google, old threads on the RR forum and motorola capacity plus sales powerpoints and PDFs, but I think I finally understand how Capacity Plus handles channel grants. (I will compile all these various links together and post them here for reference tomorrow.)

A Capacity Plus system has no system controller (which LTR, MPT1327, P25 etc all have), it is simply a network of up to six voice/data repeaters connected together (with the option for additional data only repeaters).

Unlike LTR, there is no home channel. When all radios are in an idle condition, they all sit on what's called a "Rest Channel." (We'll call it repeater A.) The Rest Channel can be identified by the rapid "Beacon" data burst approximately every 2 seconds I mentioned in the previous thread. When a user keys up a talk group, every other idle radio NOT on that talkgroup is told by repeater A to go to a NEW rest channel (on a different frequency called repeater B for this example), and only the originating radio and the other radios on its same talkgroup are allowed to stay. Then repeater A "converts" itself from being a rest channel to being a voice channel. When the voice transmission is complete, repeater A tells all radios on that talkgroup to return to an idle condition and join all the other idle radios on the new rest channel, repeater B.

Presumably there can only be one rest channel at a time, so if while repeater A was busy with group 1 and group 2 wanted to key up repeater B, the remaining idle radios would be sent to repeater C (and everyone would eventually end up there upon termination of the voice transmissions on A and B). Further assuming, if a third group tried to key up C but there was no extra repeater available a system busy condition would eventually result. Obviously multiply the preceding example by 2 to account for both TDMA voice channels per repeater.

I believe we have been looking in the wrong place (the CSBK frames during a voice call). My GUESSES:

1.) That channel grants are controlled by CSBK data frames preceding every voice call. There could be multiple types. Maybe one to ACK the request for a channel grant, one to assign all non-participating idle radios to change to the LCN of the new rest channel, and one to indicate the conversion from rest channel to voice channel (talk permit).

2.) Voice calls must be followed by CSBK data frames telling the radios to return to an idle condition and switch to the LCN of the new rest channel (where all the other idle radios were sent).

Finally, one of the documents did say late group entry was supported - for instance if the user turned their radio on after the start of a voice group call, they could still receive the end of it even though they missed the initial data frames. I have a feeling the interleaved CSBK data during the voice call (what we have been looking at) helps accomplish this, but I'm not sure how.

I hope the above interpretation was somewhat logical and helpful.

 

[inigo88]

List of references:

Forum threads:

There is no control channel on a MOTOTRBO Capacity Plus trunked system. However, there is a 'rest' channel. This 'rest' channel changes after each transmission of a subscriber unit. The 'rest' channel will key up every 'x' number of seconds (for example, every 2 seconds.....this is programmable in system).

There is a control channel for a MOTOTRBO Connect Plus trunked system. However, the control channel never changes. Next year, there will be a firmware upgrade that will allow the control channel to switch between to repeaters, but no more than two.

In MOTOTRBO in Windsor?

I need a price on a mototrbo system: Good explanations of capacity & connect plus and talk interrupt.

I see a lot of misunderstandings concerning Mototrbo trunking. It is a different format than LTR. LTR is a home chan. based system, talk groups look for a pre-programmed home chan. for system access . Mototrbo uses a rest chan./digital slot for system access. The radio sits on an un-occupied random chan/slot and waits for activity. If the activity is not intended for it's talk group the radio moves to the next open chan/slot. I does not use a system controller either like privacy plus either, so all channels can be used for activity. I know this is basic, but I hope it gives a better picture of the system.

MOTOTRBO Trunking

Outside sources:

Telcomsa.es: Mototrbo Capacity Plus Overview v.4

Capacity Plus Overview (different version)

Mototrbo version R1.6a training: Extensive details on v1.6 Tx interrupt, Dynamic Mixed Mode and Emergency Call features. Numerous CPS screen shots.

SmartPTT Console Tutorial: Capacity Plus: Tutorial for setting up SmartPTT dispatch console on capacity plus systems. Numerous CPS screen shots.

The ETSI standards documents for DMR

 

[Jay911]

That explains why I get a "data burst" (around 5-10 seconds long) on some TRBO channels, I think... nowhere near every 2 seconds (more like every 2 minutes or less frequent). Thanks.

Does every channel in a Capacity Plus system have the 'rest burst'?

 

[inigo88]

It looks like, according to some of the CPS screenshots from those attached documents, that the rest channel "beacon" burst is user programmable. So it could be every 2 seconds, 5-10, etc depending on the setting.

On my local Capacity Plus system only the rest channel has the burst, and the other 4 frequencies are silent. The rest channel rotates between repeaters, and with it so does the beacon burst. So sometimes I notice it on one frequency and other times it will be on a different one.

 

[inigo88]

I have been looking back in my logs trying to better understand the chronology of data messages sent before, during and after a voice call. Here is what I have been able to come up with, presented for everyone's analysis:

Starts with alternating CSBKO=62 + FID=16:

1110010000000000000000000000000000000000000000000000000000000000
1100010000000000000000000000000000000000000000000000000000000000

(etc...etc...)

Then switches to CSKBO=62 + FID=16:

1100001100010000110010010000000000000000000000000000000000000000

AND MIXED IN WITH IT:
Unknown Full Link Control LC : FLCO=4 + FID=16:

00000000000000000000000011001001000000110000010010001010

Continues with CSBKO=62 FID=16: (same as above)

1100001100010000110010010000000000000000000000000000000000000000

Voice Call starts, [Group: 201, Source: 1162], above CSBKO data frame continues.

When the voice call ends, we get CSBKO=62 + FID=15:

1100001100000000000000000000000000000000000000000000000000000000
1110001100000000000000000000000000000000000000000000000000000000

(...Alternating... Notice how this time bit 7 is set unlike before.)

Then Unknown Full Link Control LC : FLCO=4 + FID=16:

00000000000000000000000011001001000001000000011100000111

Mixed with CSBKO=62 + FID=16:

1110010000100000110010010000000000000000000000000000000000000000

Then more Unknown Full Link Control LC: FLCO=4 + FID=16 (different than above):

00000000000000000000000011001001000001000000010010001010

Previous CSBKO=62 + FID=16 continues...

CALL ENDS with alternating CSBKO=62 + FID=16 data frames:

1110010000000000000000000000000000000000000000000000000000000000
1100010000000000000000000000000000000000000000000000000000000000

(...etc...)

END

I believe these "Unknown Full Link Control LC : FLCO=4 + FID=16" messages might contain the message steering idle non-participating user radios to the new rest channel, because they appear both at the beginning and the end of every voice call.

Please take a look and discuss.

 

[JRayfield]

MOTOTRBO repeaters set up in an IP Site Connect system (wide-area, ip-connected, multiple repeaters, with roaming) will also 'beacon' every so often. This is also programmable, from 10 seconds up to 600 seconds.

John Rayfield, Jr. CETma
WØPM

That explains why I get a "data burst" (around 5-10 seconds long) on some TRBO channels, I think... nowhere near every 2 seconds (more like every 2 minutes or less frequent). Thanks.

Does every channel in a Capacity Plus system have the 'rest burst'?

 

[inigo88]

Hi John, thank you for joining this discussion! As you can see, while I was researching google on capacity and connect plus your posts here came up quite often and were extremely helpful.

Edit: Was my explanation of capacity plus and how it assigns rest channels somewhat correct? I did the best I could given the material I read online, but I don't have the equipment on hand to play with like you do.

Regards,

Inigo

 

[inigo88]

Well clearly the "Unknown Full Link Control LC: FLCO=4 + FID=16" messages contain both the Group ID and source radio IDs!

Unknown Full Link Control LC : FLCO=4 + FID=16:

00000000000000000000000011001001000000110000010010001010

Analysis:

000000000000000000000000 [B]11001001[/B] 000000[U]1100[/U]000 [B]10010001010[/B]
                             |                        |
                        Group 201               RID: 1162

I am curious if the underlined four bits could be a Logical Channel Number (for either the current voice channel or assigned rest channel). If it is an LCN, 1100 (binary) = 12 (decimal).

Then Unknown Full Link Control LC : FLCO=4 + FID=16:

00000000000000000000000011001001000001000000011100000111

Analysis:

000000000000000000000000 [B]11001001[/B] 00000[U]1000[/U]0000 [B]11100000111[/B]
                             |                        |
                        Group 201               RID: 1799

In this case, 1000 (binary) = 8 (decimal).

In the third "FLCO=4 & FID=16" message in the preceding post, there is again Group ID 201 and Source ID 1162, with the decimal 8 in between (in the first example it was 12). Both Source ID 1162 and 1799 were confirmed independently as legitimate radio ID's from elsewhere in the logs.

It's also pretty safe to say that the beginning and end of a call is denoted with the alternating CSBKO=62 + FID=16 message starting with:

111001...
110001...

Please feel free to post some guesses or ideas. I feel like I'm having a conversation with myself.

 

[IanWraith]

Superb work Inigo88 !!! I was hoping to have a chance to look into this over the weekend but sadly didn't have the time. I shall integrate your information into DMRDecode and hopefully release a new build later this week.

I will also have a look in my logs to see if that casts any light on the mystery 4 bits.

Once again many thanks.

Ian

 

[Forts]

I don't know if it will be of any help or not, but there is some nice info on Capacity Plus in this document...

http://www.smartptt.com/files/documents_eng/SmartPTT Tutorial - MotoTRBO Capacity Plus.pdf

 

[Forts]

Was looking thru a few of my capture logs trying to see if anything I have jives with what you guys are seeing.

I was logging a Capacity Plus system utilizing 3 Cap + frequencies, and found this entry:

Unknown Full Link Control LC : FLCO=4 + FID=16:
 0000000000000000000000000000 [b]0100[/b] 000000[b]11[/b] [b]0000111110111000[/b]
                                |                     |
                             Group 4           Radio 4024

According to the logs, Radio 4024 was seen on Group 4... which matches.

Now if we are thinking that 0000011 might be a channel assignment, then in my case this makes sense too. As I mentioned above, this system uses 3 Capacity Plus frequencies. When these are programmed into the radios they are assigned ID's (frequency 1 gets ID 1&2, frequency 2 gets 3&4, frequency 3 gets 5&6 etc). From the CPS:

Now in this case I know the transmission took place on the 2nd frequency, which would be assigned ID's 3&4.... So the 3 matches the 00000011 above. Perhaps on a given frequency, the assigned ID represents the timeslot? In my example ID 3 would be timeslot 1, while ID 4 would be timeslot 2. Now I could be completely off track here, but it somewhat sort of makes sense. I'll see if I can do more logging on this system so I can have a few more samples to analyze.

In inago88's example above, 1000b/8d could be referring to the 4th frequency in use for that system, possibly timeslot 2?

 

[Jay911]

I've further assessed a local multi-frequency TRBO installation and think it might be Capacity Plus after all. I'm running DMR build 32 on one of the freqs right now and will post all the logs once I capture some input from all of them.

 

[inigo88]

Forts, I am so glad our FLCO data is looking consistent! I'm also glad you found the SmartPTT pdf with the CPS screenshots, I cited that link in the references at the beginning of this thread. Your interpretation that the trunking system looks at the two TDMA time slots as separate LCN IDs is absolutely right! It's explained in the following quote from the DMRDecode thread (before we spun off):

From an infrastructure point of view, each repeater is issued a pair of channel numbers ( CH 1,2 or 3,4 or 7,8). So here is how I have most of my systems setup:


Rptr A 451.000mhz gets assigned as "CH 1,2"
Rptr B 452.000mhz gets assigned as "CH 3,4"
Rptr C 453.000mhz gets assigned as "CH 5,6"
etc all the way up to 6 voice and 6 data repeaters (total of 12 voice and 12 data simultaneous slots) for 1 single site cap plus system.

I will see if I can get some programming screen shots. It is very similiar to LTR except Cap Plus fails gracefully. I have a demo system at my disposal that we can take captures of it will help (using known system parameters). See the Mototrbo System Planner for more detail. TT

Also I noticed we broke up our FLCO=4 PDU bits differently, but that brings up a very good point: I arbitrarily decided where to put spaces in the FLCO=4 PDU using my best guess (and I imagine so did you). So my four bits for the hypothetical LCN field between the group and source ID fields could actually be more or less bits.

I think the best way to determine the exact dimensions of this field would be by process of elimination, by looking at the maximum value the group ID and source ID can be, and then figuring out how many bits of binary code this will take up. The unknown field will then be whatever is left in between.

This is probably a question best directed at Ian, but does anyone else know if the ETSI DMR standards specify the maximum values the group and source IDs can take up? (For instance Motorola systems sometimes use 65535, the highest possible talkgroup value, as the CW ID.)

I wrote this from an iPhone so please pardon grammatical errors.

 

[Jay911]

I ran DMR Decode 32 on a batch of frequencies nearby my location overnight tonight. I don't know if they form a trunking system, but I think so - the burst every 2 seconds is there as described above, and I also see some different data bursts that come by about every 60 seconds, almost like an idle stream. There are a few voice calls in here as well. I wrote a bunch of small files as I ran from frequency to frequency, then concatenated them all together, arranged by frequency.

Some of the frequencies are stronger than others, despite the fact that they are all coming from the same site around 5 blocks away from me. Frequencies 452.4375, 462.2375, and 462.5375 are supposed to be all at the same ERP (~5W) so I was thinking they're supposed to be a system together. 463.9375 is a little weird - it is used on this site as well as at a more distant site (around 5-10 miles away), both owned by the same company and both presumably TRBO. It also seems to be in use in analog mode somewhere else nearby.

Anyway, here's the data - take from it what you will!

 

[Forts]

As per the CPS, the maximum value for a Capacity Plus Group ID is 254, Private ID is 65535, and All Call is fixed at 255. For non-Capacity Plus the max Group or Private ID is 16776415, with All Call using a fixed value of 16777215.

So it looks like for Capacity Plus it's 8 bits for the Group ID and 16 for the Radio ID.

 

[inigo88]

Thank you Forts!!!

000000000000000000000000 [B]11001001[/B] [u]00000011[/u] [B]0000010010001010[/B]
                             |                        |
                        Group 201               RID: 1162

Here is one of the previous PDUs I quoted with the proper formatting. It looks like the LCN ID is 3 (1st time slot of frequency 2).

The LCN ID in the other example is 100b = 4d.

On the assumption that you can have 6 cap+ frequencies x 2 slots per freq, the LCN IDs should be <= 12. So far so good!!!

 

[Forts]

I ran DMR Decode 32 on a batch of frequencies nearby my location overnight tonight. I don't know if they form a trunking system, but I think so - the burst every 2 seconds is there as described above, and I also see some different data bursts that come by about every 60 seconds, almost like an idle stream. There are a few voice calls in here as well. I wrote a bunch of small files as I ran from frequency to frequency, then concatenated them all together, arranged by frequency.

Some of the frequencies are stronger than others, despite the fact that they are all coming from the same site around 5 blocks away from me. Frequencies 452.4375, 462.2375, and 462.5375 are supposed to be all at the same ERP (~5W) so I was thinking they're supposed to be a system together. 463.9375 is a little weird - it is used on this site as well as at a more distant site (around 5-10 miles away), both owned by the same company and both presumably TRBO. It also seems to be in use in analog mode somewhere else nearby.

Anyway, here's the data - take from it what you will!

It looks like 462.5375 and 463.9375 belong together (with 463.9375 using LCN's 1 & 2, and 462.5375 using LCN's 3 & 4). There wasn't enough data logged on the other channels to see where they fit in. Interesting that 463.9375 is used a short distance away... some sort of simulcast setup perhaps?

 

[TampaTyron]

...............and thats how we decode capacity plus! TT

 

[Jay911]

It looks like 462.5375 and 463.9375 belong together (with 463.9375 using LCN's 1 & 2, and 462.5375 using LCN's 3 & 4). There wasn't enough data logged on the other channels to see where they fit in. Interesting that 463.9375 is used a short distance away... some sort of simulcast setup perhaps?

That's the downside of monitoring from 6pm to 6am instead of a normal human's hours.

I have some spare time coming up later this week in which it's my intent to go to both sites and try to do some better, but shorter, logging.