TRBO voting ideas

[iamjason]

Sorry if I'm replicating an older post (I've searched and haven't found anything similar so here goes).

I've been throwing around ideas in my head to get voting at our TRBO site short of having to buy several other XPR8400s or MTR3000s just to use as receivers which is cost prohibitive for our operation. This is a Ham site so we have some leeway to experiment.

Has anyone put some thought into or have tried the following?:

I'm thinking of setting up a voting system separate from the repeater that would pass digital signals transparently end-to-end. The input side of this system would be the desired repeater INPUT frequency. And the output side would be a link back to the repeater with the repeater itself being set to receive on the link frequency and of course, transmit on the desired OUTPUT frequency.


Example:

RX 443.000 / TX 448.000

INPUT: Receiver (443.000) / Link Radio (902.000) --------[link]------> Link Radio (902.000) / Voter / Link Radio (430.000) -------[link]-----> Repeater (IN 430.000 / OUT 448.000) :OUTPUT



Also, does anyone know how the receive radio in the XPR repeaters communicates with the controller? Is it just a dumb receiver passing audio to the controller, or is there processing, etc that is done in the receiver before it is passed to the controller? If the receiver is just passing a generic signal, I wonder if it might be possible to remove the receiver and somehow connect the voter directly to the controller.

It goes without saying, that for anyone familiar with cheap MaxTrac P25 repeaters this would of course be a compromise, but its the only way I can see that would pass talkgroups, IDs, etc intact. And of course, there are times where tying up the repeater with a horrible signal is WORSE than not getting in at all so this all may be moot if signal quality across the system turns out to be horrible. Plus there's the issue of getting an analog voter to vote on a digital signal.

 

[moetorola]

I don't think what you are wanting to do would work. At least like the poor man's P25 repeater maxtracs. It is TDMA.

I believe you will have to go the TRBO route to do what you are wanting.

I am no TRBO expert though.

 

[iamjason]

True, I hadn't even thought of the TDMA issue. Theoretically you could pass TDMA providing both timeslots are going in the same direction - i.e. TS1/TS2 are TX/TX or RX/RX, but not TX/RX or vice-versa which is bound to happen.

Damn. I'll stack this up to having beer for breakfast.

 

[com501]

Yup. This WILL NOT work. You need to have a digital path 100% of the way to establish sync for the voting receiver(s) and for all the other parameters that are necessary for 2-slot TDMA to work. Even the receive RSSI is transmitted digitally as part of the receive packet from the voting receiver.

 

[RFI-EMI-GUY]

Mototrbo IPSC has digital voting now as a feature. I have no experience with it.

 

[Avery Johannssenn]

Mototrbo IPSC has digital voting now as a feature. I have no experience with it.

Too bad the XPR mobiles don't have any ethernet connectivity. They're a hell of a lot cheaper than an MTR3000.

 

[K5TLF]

Yup. This WILL NOT work. You need to have a digital path 100% of the way to establish sync for the voting receiver(s) and for all the other parameters that are necessary for 2-slot TDMA to work. Even the receive RSSI is transmitted digitally as part of the receive packet from the voting receiver.

This...
And not to mention that the noise on 900 when converted to digital sounds like absolute shit...

 

[Mars]

I'll throw my two cents in, too.

No one should waste time thinking about doing this unless they have a private LAN/microwave link between all of the sites/voters. Latency must be extremely low; 99.99999% of Internet connections aren't reliable enough.

In order for the voting functionality to be working properly, the one way network delay between the
repeater and any of its receivers must be less or equal to 40 milliseconds. Additionally, the network
asymmetry between the repeater and any of its receivers must be less or equal to 12 milliseconds.
The network asymmetry is the absolute value of the time difference for an IP packet to travel from
the repeater to the receiver, and from the receiver to the repeater. This applies to all system
configurations. Since the distance between the repeater and receiver is normally less or equal to
90 miles (approximately 145 kilometers), most of the business grade IP networks are able to meet
this 40 milliseconds per 12 milliseconds network requirement.

 

[RFI-EMI-GUY]

Mars.

The Mototrbo voting could be used to provide reciever diversity gain using two co located stations. That would provide a talk back benefit for portables, exploit multipath and would avoid the hassle of jitter and latency otherwise resulting from constructing a wide area network.

 

[Mars]

Mars.

The Mototrbo voting could be used to provide reciever diversity gain using two co located stations. That would provide a talk back benefit for portables, exploit multipath and would avoid the hassle of jitter and latency otherwise resulting from constructing a wide area network.

Agreed 100%.

But no way this can happen if using the Internet to link sites/voters together. It's too latent. Need to go with a private LAN or wireless LAN (using microwave) instead.

 

[RFI-EMI-GUY]

Which is why I said "Co Located Stations". Traditionally LMR has used site diversity to improve talkback, but that requires having access to multiple towers (leases), phone or network connectivity. But having two receivers on the same tower at different heights balances talk back from portables. W. C. Y. Lee wrote some early reports on receiver diversity. For a tower, two receivers with vertical separation, a rooftop, three receivers with horizontal separation to ensure the signals are uncorrelated.