MotoTRBO portabel digital simplex between 2 moving cars

[maelv]

Hello all

i am new her on batlabs but i have read all i could find on the batboard about mototrbo

but is there any one here ther have tried the subject ?
because i have read about an other digital system that uses the same ambe vocoder as the mototrbo (the jarl D-STAR standard) and in digital simplex between 2 moving cars the audio is almost unread able because of the signals multipathing and FEC in the radios

and ther for i would like to know if any one have tried to test that with the mototrbo..to see if it to has the same problems?
i am currently looking to buy the euro version of the XPR6550 as a live in europe..
the model name here is DP3601

Maelv

[dxon2m]

There's certainly gonna be multipath problem but that also depends on a lot of other variables...ie. how fast are you moving; how far apart are two cars; what power level are you using; what is the environment? urban/forest or desert?

[wavetar]

I haven't tested it specifically for the TRBO radios, but I know with the Motorola Astro radios using the IMBE vocoder movement isn't an issue until you approach 160+ miles an hour. Even then, there's a programmable setting (high deviation tx under trunking system options) to allow them to work better at high speeds, which leads me to believe there's more to it than the vocoder itself. Perhaps the Motorola radios are better able to handle it than the Icoms? It would be interesting to try out.

Todd

[maelv]

Thanks for the infomation

i do also think that because the mototrbo radios uses TDMA and the speech is very compressed that it might not be a problem but if any one have the gear/time to make a test it would be nice to know how it works.

and maybe because the trbo have a better error correction system..? but i dont know

do any body have new information about the security i trbo... is there comming encrypting to them ?

Maelv

[fireradio]

From what I've seen playing with some demo units, there are no encryption options with the TRBO radios.

[mr.syntrx]

Nor is there any provision in the ETSI standards for crypto.

Unless Motorola implements something strange and proprietary in that regard, it's not going to happen soon.

[Wowbagger]

The vocoder is not going to do ANYTHING to your resistance to multipath - that is entirely a function of the modulation scheme. C4FM as used in APCO-25 has problems with multipath because any timing error in when you sample the signal will lead to a MASSIVE increase in inter-symbol interference (ISI), leading to decode errors that the foward error correction (which, again, has NOTHING to do with the vocoder) must correct - and every bit error FEC has to correct due to mis-slicing is that much less resistance to errors due to low signal level.

There are modulation schemes in which the "eye" of the signal is more open, meaning that more timing error can be tolerated before you get into slicing errors - however that comes at the price of increased occupied bandwidth for the signal.

The probable issue here isn't one of the physical layer transport, but rather the fact that in order to get the very high compression ratios needed to make these schemes work, the vocoder has to make a lot of assumptions about the signal being encoded - chief among them being that the signal is composed ONLY of voice data, and nothing else. Add a bunch of other noise and the vocoder starts to make invalid assumptions about what is important and what can be thrown away. I would suspect the real issue here is not bit error rate, but rather the noise in the vehicles - use a noise canceling microphone or a throat mike, or even a boom mike very near to the mouth with the sensitivity way down, and that will probably make all the difference in the world.

Again, people - please don't conflate the vocoder, which is SOLELY responsible for turning analog wiggles of the microphone into a bit stream, with the protocol, which sets the rules for how those bits get sent over the air, with the modulation layer, which converts the bit stream into RF wiggles. IMBE and AMBE are not modulation schemes nor are they protocols. You can do APCO-25 protocol over modulation schemes that are not APCO-25 C4FM, and you could in theory use APCO-25 C4FM to carry any data stream that can be represented as 9600 bits per second.

[fireradio]

Here's a question: Would there be any reason a split antenna system (2 antennas to 1 repeater) would cause any complications while operating in TRBO digital mode?

[Wowbagger]

Here's a question: Would there be any reason a split antenna system (2 antennas to 1 repeater) would cause any complications while operating in TRBO digital mode?

I assume you mean one antenna for RX, one for TX - as long as they are properly isolated from one another so the TX does not desense the RX, I don't see why such a system would have any problems that a duplexed single antenna wouldn't have.

[fireradio]

Here's a question: Would there be any reason a split antenna system (2 antennas to 1 repeater) would cause any complications while operating in TRBO digital mode?

I assume you mean one antenna for RX, one for TX - as long as they are properly isolated from one another so the TX does not desense the RX, I don't see why such a system would have any problems that a duplexed single antenna wouldn't have.

Not quite - I'll be setting up a repeater in an 8 story building, so my local dealer recommended a split system that uses 2 antennas -- each capable of transmitting and receiving, but splitting the total output power to each antenna in half. This would allow one antenna to cover the upper floors, another to cover the bottom. See http://en.wikipedia.org/wiki/Distributed_Antenna_System

[mr.syntrx]

I'm guessing it could possibly cause problems, if there's enough of a phase difference in the signals from each antenna as they arrive at the receiver to cause problems demodulating the signal (a bit like ghosting on TV.)

I'd look at a leaky feeder system instead.

[libuff]

I haven't tested it specifically for the TRBO radios, but I know with the Motorola Astro radios using the IMBE vocoder movement isn't an issue until you approach 160+ miles an hour. Even then, there's a programmable setting (high deviation tx under trunking system options) to allow them to work better at high speeds, which leads me to believe there's more to it than the vocoder itself. Perhaps the Motorola radios are better able to handle it than the Icoms? It would be interesting to try out.

Todd

sorry, i have to ask.. what happens at 160mph? either let me know, or im going to try!

really tho, Trbo is the same as the icom d-star ham radios? thats pretty cheezetacular.. are they interoperable?

[mr.syntrx]

really tho, Trbo is the same as the icom d-star ham radios? thats pretty cheezetacular.. are they interoperable?

Other than the vocoder, D-Star and MOTOTRBO have nothing in common.

As Wowbagger pointed out, the vocoder is only a very small part of the overall system.

[Wowbagger]

Not quite - I'll be setting up a repeater in an 8 story building, so my local dealer recommended a split system that uses 2 antennas -- each capable of transmitting and receiving, but splitting the total output power to each antenna in half. This would allow one antenna to cover the upper floors, another to cover the bottom. See http://en.wikipedia.org/wiki/Distributed_Antenna_System

Well, the issue will be in the overlap zone between the two antennas, where the signal strength from each antenna is close enough to the other signal that the radio cannot capture only one.

This is why Motorola came up with their linear simulcast system - as I've said in another post the "eyes" of the signal, the times during which the carrier is at the proper frequency to be decoded into the correct symbol, are larger, so during a longer period of time the two signals will be at the same frequency and will thus be seen as one signal by the radio.

I don't recall the specifics of the MotoTRBO modulation - I'm not writing the slicer for that, one of the DSP guys is, so I don't recall if the eyes are wide open like LSM or are more closed like APCO-25.

And as to the question of "Is TRBO like D-STAR" - are you like a chimpanzee? You are both made of meat, you have similar skeletal structure, you have a large amount of DNA in common - does that mean you will "interoperate" with a chimp, and produce offspring?

Both use 4 level FM, both use AMBE, but the specifics of the modulation, the protocol format, and such are different.

[maelv]

Hello every body thanks for the reply

i now know a littel bit more about mototrbo

i am now waiting for at quotation from my dealer

Thanks

Maelv

[wavetar]

I haven't tested it specifically for the TRBO radios, but I know with the Motorola Astro radios using the IMBE vocoder movement isn't an issue until you approach 160+ miles an hour. Even then, there's a programmable setting (high deviation tx under trunking system options) to allow them to work better at high speeds, which leads me to believe there's more to it than the vocoder itself. Perhaps the Motorola radios are better able to handle it than the Icoms? It would be interesting to try out.

Todd

sorry, i have to ask.. what happens at 160mph? either let me know, or im going to try!

Good question...I don't know. It's just what the help files say about that particular setting. I assume it was meant for use in things like helicopters which could surpass that speed routinely.

Todd

[libuff]

Perhaps doppler shift? but how common is that, and would changing a deviation make the doppler shift more acceptable? how bad is the doppler shift in the MHZ range anyway? its noticable in the HZ range with audible noise, but is it that different in the higher frequencies? this is about the extent of my knowledge, from HS physics and stuff..

[Wowbagger]

Perhaps doppler shift? but how common is that, and would changing a deviation make the doppler shift more acceptable? how bad is the doppler shift in the MHZ range anyway? its noticable in the HZ range with audible noise, but is it that different in the higher frequencies? this is about the extent of my knowledge, from HS physics and stuff..

OK, let's do the math. The symbol deviations for P25 are +/- 600 Hz, +/- 1800 Hz - that means that at symbol time the carrier should be either 600 Hz or 1800 Hz away from nominal center frequency, either above or below. Now, the occupied bandwidth for normal P25 is about 12.5 kHz ( for check: 2*1800 Hz deviation + 2*4800 Hz symbol rate = 13.2 kHz, but the shaping filters specified in APCO bring that into 12.5 kHz at the expense of a slightly higher inter-symbol interference level).

So, given 25 kHz channel spacing, you can move about 6 kHz before you really start to interfere with the adjacent channel. At 12.5 kHz spacing you will be into the adjacent channel almost immediately, but FM capture effect is your friend here, so you have about a kHz or so. A good data slicer can deal with an error of 6 kHz or more (I know, as the slicer I wrote can do exactly that), so the real limiting factor on the receiver is the narrowness of the IF filter - a 12.5 kHz brick wall 255 tap FIR filter like we use will start to distort the signal at about a kHz.

APCO spec for frequency error is .1 PPM - so for an 800MHz radio the specified carrier frequency error should be less than 80 Hz to be in spec.

OK, so now we have some context into which to put the Doppler calculations. Doppler effect consists of 2 effects based upon the velocity vector between the transmitter and reciever: that which is caused by the radial component (motion along the line between TX and RX) and relativistic effects of the tangential motion (which can be ignored for any speeds we will see on this planet). The equation is
Fchange = Freal*V/c. Assume V=160 mile/hour and Freal=800MHz. Then the observed frequency is 191 Hz off nominal, or about twice the specified maximum carrier frequency. It is much less than the symbol deviation, much less than what would be required to bang against the side of a tight IF, much less than what a good data slicer can track.

OK, what about timing error? A symbol time is 1/4800 of a second, or about 200 microseconds. c is about a nanosecond a foot. In one symbol time, an object moving 160 miles/hour will move about .05 of a symbol time - again, not enough to cause a significant error in the slicer.

I'd say the real issue is not the speed per se, but rather the implications of an object which can travel at those speeds under normal conditions.