dslreports logo
 
    All Forums Hot Topics Gallery
spc
uniqs
12
viasatguy
Premium Member
join:2002-06-11
Carlsbad, CA

viasatguy to DrStrangLov

Premium Member

to DrStrangLov

Re: ViaSat-1:: Mobile Internet

It is going to perform better than the 75 cm (2.5 ft) size antenna by about 4 dB - basically 20 times the log of the diameter ratio.

But there will certainly be weather conditions that severe enough to cause outages some of the time. The system has adaptive coding and modulation so the link rate will be adjusted as needed to keep the link alive as long as possible and it now has 4 extra dB to play with.

The real benefit though is the increased throughput in clear sky conditions. The 1.2 m antenna should allow operation at a higher data rate most of the time, and for SNG operators this is a great thing because they can transmit their news files faster, or send a higher quality live video feed, etc.

grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY

grohgreg

Member

said by viasatguy:

The real benefit though is the increased throughput in clear sky conditions.

I'm not buyin' that. Increasing dish diameter is generally consistent with gain. But the connection is digital, so it's either on or off. That means - under clear skies - it's not going to work any faster than the gateway will respond. The additional gain is really most beneficial in adverse weather conditions,by extending the amount of time before the connection to the gateway is dropped.

//greg//

dbirdman
MVM
join:2003-07-07
usa

dbirdman

MVM

said by grohgreg:

I'm not buyin' that. Increasing dish diameter is generally consistent with gain. But the connection is digital, so it's either on or off. That means - under clear skies - it's not going to work any faster than the gateway will respond. The additional gain is really most beneficial in adverse weather conditions,by extending the amount of time before the connection to the gateway is dropped.

//greg//

Greg, it sounds like you are thinking of a .74-meter account that puts a 1.2-meter in on the same account. In that case, speeds in clear weather will be identical.

ViaSatGuy is talking about an SNG account, which could be paying for 50Mbps of bandwidth or more. It takes a lot of power to push this kind of bandwidth, and the options are more powerful transmitters and bigger dishes.

I've seen 200-watt transmitters attached to 1.2-meter dishes for Ku-Band SNG applications. The transmitter is typically water-cooled and located inside the vehicle with a flexible waveguide out to the feedhorn.

The world expects HD video from every remote feed, and that's what it takes. Big SNG trucks of course have 1.8 or 2.4 meter dishes, but there are a lot of news vans that can't handle that size.

grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY

4 edits

grohgreg

Member

Nope, I'm stickin' to my story. Transmitter output has nothing to do with antenna gain. The average difference in gain between a .74 and a 1.2m is ~4.5dB. On a linear scale, that represents an average 11% increase. And antenna gain has nothing to do with throughput speed either, consumer or commercial. On the RX side, increased gain represents enhanced survivability in adverse weather conditions. It only affects throughput by holding a signal longer before the FEC process kicks in.

On the TX side, it's all about power. Yes, higher antenna gain increases EIRP proportionately. But in the example you give, it's all about power per bit. The more bits you're pushing, the more composite power is required to maintain the same power per bit. Hence a 2 watt transmitter for a 500kb bitstream, and a 200 watt transmitter for a 50mb stream. In the end, they both use the same amount of power per bit. Then factor in antenna gain, and the 1.2m will simply give the bitstream a little more (effective) power per bit - which again - goes to survivability.

//greg//
viasatguy
Premium Member
join:2002-06-11
Carlsbad, CA

viasatguy

Premium Member

The actual increase in max EIRP would go from 48.8 dBW to 54.3 dBW (the differene in the max EIRP of the FCC filings between the two antennas). This is a change from 75857.8 kW to 269153.5 kW -roughly a 350% increase.

Because the ViaSat system has adaptive coding and modulation in addition to uplink power control, it can dynamically change the modulation and forward error correction coding, and/or increase the symbol rate if conditions permit.

So, if there is enough extra power available at the terminal to go to the next higher symbol rate, the system would make the jump. Power control is then used to make sure that only enough power is used to close the link at the given symbol rate and mod/code point. A static excess margin is not transmitted - the system is constantly adjusting the power to keep the link at the desired Es/No.

grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY

grohgreg

Member

Like I said, it's all about power. Antenna gain change doesn't change transmitted power, it changes EIRP. If you're uncomfortable with "power per bit", let's go with "energy per symbol". They're essentially interchangeable phrases.

But perhaps a reminder is in order, Dr Strange ostensibly opened this topic to discuss mobile internet - not SNG

//greg//
viasatguy
Premium Member
join:2002-06-11
Carlsbad, CA

viasatguy

Premium Member

I think we may be in agreement but perhaps saying it a different way.

In both cases the TRIA assembly has a maximum power of 2.8 W (4.47 dBW). So the transmitted power into the antenna does not change in either case - which I think was was your point. If so, we are in agreement so far.

The 75 cm antenna has a gain of 44.5 dBi (gain relative to an isotropic source) at 30 GHz. The 1.2 m antenna has a gain of 49.8 dBi at 30 GHz.

So what does change is the effective radiated power in the direction of the satellite.

If the TRIAs on both antennas are transmitting at max power then the satellite will perceive the signal from the 1.2 m antenna as being 5.3 dB stronger than the 75 cm antenna.

Let's say both are transmitting a 5 MBd (5 mega baud, or 5 mega symbols per second) carrier with QPSK modulation and Rate 3/4 FEC. The effective data rate is 7.5 Mbit/s.

The effective isotropically radiated energy per bit is 10.52 mW/bit for the 75 cm antenna and 35.64 mW/bit for the 1.2 m antenna. So the energy per bit is higher for the 1.2 m antenna - and as you said, we could have just as easily expressed this in terms of energy per symbol (which in fact is what we often do).

Now, let's look at a simplified clear sky link budget for each. The EIRP of the 75 cm antenna is 48.8 dBW and the EIRP of the 1.2 m is 54.3 dBW. The path loss is 213.4 dB and the satellite G/T is 22 dB/K. For grins, let's also lump in another 5 dB of generic other losses.

The resulting C/No for the 75 cm antenna is 81 dB-Hz.
The resulting C/No for the 1.2 m antenna is 86.5 dB-Hz.

The energy per bit to noise ratio (Eb/No) is 12.3 dB for the 75 cm antenna and 17.7 dB for the 1.2 m antenna.

It's not too hard to find a R 3/4 decoder that works just fine at a 5 dB Eb/No, so we have lots of margin over what we need to properly demodulate the signal.

What do we do with that? Well, we could just leave it there as margin against weather. That would be kind of wasteful though since it's not going to be raining most of the time and most of the time we would be just throwing power and bits away.

So because we have a dynamic system, we adjust the symbol rate or mod/code point until the link is closed with some reasonable margin. If signal conditions change, we either make minor power corrections, or change the mod/code point and/or symbol rate until the link is happy again.

It should be pretty clear that the 1.2 m antenna has capability to produce a higher C/No or Eb/No at the and hence offers the ability to either transmit at a higher rate, or be more resistant to rain fade, or even operate further out toward the edge of a given beam contour.

grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY

2 edits

grohgreg

Member

No problem, I have a firm grasp of the dynamics of adaptive systems. And in case we have a terminology problem here, my acronym EIRP is universally understood among RF specialists as "Effective Isotropic Radiated Power". And I think you understood that. But your satellites don't "perceive" anything. Given that you're not using onboard switching, their sole function is to receive a signal from Terminal A, amplify it, then relay it to Terminal B.

But for all your carefully worked out numbers, it still comes down to power per bit (or energy per symbol). I'm not sure if you use BER or Alfa Flunk, but the resultant error rate(s) is/are what drive(s)s your adaptive algorithms. What's left over is margin, and margin equates to survivability. Given the same data rate/coding and transmitter power, the increased EIRP of a larger dish - increases survivability.

//greg//
viasatguy
Premium Member
join:2002-06-11
Carlsbad, CA

viasatguy

Premium Member

So perhaps we're all in agreement now.

I agree with the definition of EIRP.

"Given the same data rate/coding and transmitter power, the increased EIRP of a larger dish - increases survivability.
"
I agree with this as well, though I might have used the phrase link availability.

Perhapls I should have said an observer at the satellite's receiver would perceive, rather than the satellite perceives, but I think you got my point.

Regarding mobile Internet. The first post in this threat talked about mobile, fly-away, and vehicle-mounted antennas.

Mobile in the strictest sense is on-the-move, whereas communication from SNG trucks, RVs, and Fly-away terminals would typically be a temporary fixed application.

ViaSat is doing both. The JetBlue in-flight service would be an example of a mobile application and the SNG van (could just as well be an RV) and the fly-away terminal would be examples of a temporary fixed application.

Both are often lumped into the general mobility term, but in any case, it doesn't matter to the link whether the bits being transmitted are for Internet access or digitally encoded HD video.