: Understanding OTA DTV Broadcasting Technology in Canada
2005-05-21, 03:24 PM
Conventional analog TV requires whopping amounts of power for a given area (several hundred KW, or maybe even a MW or two!). On the other hand, a 100 Watt VHF/FM radio transciever with an excellent antenna design and location can cover the exact same area with voice-only communications, such as for a courier or taxi company. Now imagine if you could take the TV signal and put it into a very tight stream of ones and zeroes... it would still be more complex a signal than the voice one I mentioned, but much tighter and more cohesive than the big old analog TV signal. The outcome of years of testing and experience is that digital TV transmitters do not require anywhere near the sheer ooooomph of the old analog ones. The difference is startling at first, but when you see it with your own eyes you get it. The best examples are down in the U.S. where the power differential is quite striking, yet the area is similarly covered.
I'm not saying I'm happy with the numbers I'm seeing used in Canada, since I suspect the Canadian broadcasters, who often have stakes in satellite and cable TV services, have ulterior motives for keeping them low.
...TV signals travelling via satellite or cable TV systems undergo lots of compression and processing on their way to you, Above statement quoted from the following thread...
This is dependent on the cable company. For example there is absolutely no compression or processing of signals on Rogers (Toronto) channels 502-531. Full 19.4 Mbps bandwidth is allocated to those channels.
When people see differences in quality, it could relate to the quality of the STB itself.
It is true that Satellite must compress the signals some, due to limited bandwidth. Some cable channels and some cable companies are allocated at 3 channels per QAM, which means that roughly 13 Mbps is available, therefore there is compression of those channels.
Stampeder, I hope you don't mind these comments, but I thought that the "sweeping statement" needed clarification. If you wish, you may delete this post, perhaps changing the "sweeping statement".
2005-06-04, 04:15 PM
When people see differences in quality, it could relate to the quality of the STB itself.
Stampeder, I hope you don't mind these comments, but I thought that the "sweeping statement" needed clarification.
...Fair enough 57, but I need to clarify what is a problem for satellite and CATV providers: if a station in Vancouver simultaneously broadcasts OTA DTV while sending the same signal over fibre-optic or satellite to the CATV/DBS provider's network control centre (Calgary, Toronto, Montreal?) that signal is processed through switches and other devices that route it digitally across their network or up to a satellite and then to the subscriber's home receiver. Does it undergo compression/recompression in the process? As you say, it all depends.
But, try as they might, the providers cannot ensure that the original signal is pristine and clear of degradation or artifacts by the time it has made its full journey to the home receiver. For a Vancouver subscriber the signal's round trip can account for a visual and audible difference when compared A-B with a direct OTA signal.
Here's what I see as the problem: in my field, computer data over Wide Area Networks, its essential to have error-checking, which requires a certain amount of budgeting for time and expense. The benefit is that if something didn't arrive properly, the error is detected and the other end sends it again, but in CATV/DBS its my conjecture that signal transport exacts a tangible toll on quality because the providers cannot spend the time and money to have full error correction due to the end point user requirement for receiving the data serially in real time (on their TV screen). Streaming of serial video is a real beeyotch, especially when several layers of transport protocol are added. This "real time" issue is the same problem that has bedevilled Voice-Over-IP over the years, although good progress has been made in that technology.
While dropouts, pixellation, macroblocking, and other problems can certainly occur OTA, the signal (when good) is as clean as technically possible. When the signal is routed over vast geographic distances and must be delivered serially the options for error-checking and correction become fewer and less effective.
That's my central point, and I'm sticking with it. ;) I'll edit the FAQ to remove the references to compression but I'd like readers to see these replies so that they know that we are serious about getting our info right and that we accept and encourage debate, so thanks 57!
Perhaps it would be good to start a thread on this issue of OTA vs "uncompresed" cable....
You say there are "differences" between the signal that gets to the OTA STB (without switching, etc.) and the signal that gets to the cable STB (your conjecture). I'm not sure in this digtal age whether there are ANY differences at all in the signal.
I'd hazard a guess that there are none - ie, the bits/bytes that get to my Rogers SA8300HD from Detroit CBS are "identical" to the bits/bytes that get to the OTA STB of someone gettting Detroit CBS OTA.
This must of course be differentiated from the cable/sat providers that do compress the signal.
2005-06-04, 04:58 PM
I hear what you are saying and I have no actual evidence other than empirical A-B testing of a Bell ExpressVu top-of-the-line receiver's signal versus a top-of-the-line OTA receiver. The difference in picture and audio quality is stunning.
I wish I could route the encrypted satellite signal into the LG OTA tuner to find out definitively what differences there might be, but the LG only takes clear QAM. Without that I think its important that our readers know what they will be seeing on their couches in front of their HDTVs. My seat-of-the-pants conjecture is that in most cases OTA provides better signals that viewers will easily see.
QAM is cable. I believe Satelllite is 8PSK. Satellite IS compressed (BEV is probably around 15 Mbps (proprietary information), vs 19.4 OTA.)
2005-06-04, 05:18 PM
QAM is cable. I believe Satelllite is 8PSK. Satellite IS compressed (BEV is probably around 15 Mbps, vs 19.4 OTA.)Ooops, that's what I meant to think/say... ;) I've gone back and edited my other posts for clarity and to take into account your information.
2005-06-05, 09:40 AM
PQ of ATSC HD is severely affected by multicasting.
Checkout the "DTV PSIP Captures" page on www.dxfm.com
Some PBS channels run up to 4 SD broadcasts during the day, so the PBS demo loop is running at 10 MB/sec or so. At night, they usually go to 1 HD + 1 SD so that the HD is around 16 or 17 MB/sec (see Cinninatti + Columbus PBS examples)
Other channels run multiple SD 24/7. Example WKYC-DT Cleveland (1 HD + 3 SD) - HD is 12 MB/sec maximum.
Many other channels run local weather and traffic SD 24/7 so HD is around 16 MB/sec
Detroit PBS multicast - it's 1 SD + 1 HD. That's why they use 720P too because 720P requires less bandwidth.
I've never heard of anyone doing 3-4 SD + HD. It's usually only SD when they do that many SD, but if you have knowledge of Cleveland doing so, so be it.
2005-06-05, 03:09 PM
Repost from an original thread:
One major reason the old analog transmitters required so much power was that the video portion of the channel is AM (Amplitude Modulation). To deliver a reasonable signal to noise ratio at a fringe distance, the transmitter power had to take into account what the background electrical noise level was in the environment and the noise ceiling of vintage electronics (tubes receivers).
The sound portion of the same TV channel is FM (Frequency Modulation) which requires less power to deliver the same signal to noise ratio at a fringe distance. But the real reason that the audio uses FM and not AM is that if it were to use AM, the 60 hertz vertical scan frequency would generate noise and harmonics that would be picked up by the audio amplifier.
When you only have to transmit a series of ones and zeros by slightly changing the phase of the transmitted signal carrier wave, the power of the transmitter can be reduced enough so that at a fringe distance, the TV tuner need only detect a slight change in phase in a fairly weak carrier signal.
Unfortunately there will always be some random noise (spikes) that will cause the receiver to momentarily lose track of the incoming signal and create that confetti look on your HD screen.
If the only way to deliver HD was over the air, then I think extra bandwidth would be allocated to the channel to provide some sort of error checking and recovery like is used in our computer memories and compact disks.
Because so much of HD is and will be delivered by other than OTA, there has not been the same attention to quality (more transmitter power/error correction data) that was present when television was first introduced and a rooftop antenna was the only method of reception.
2005-06-05, 03:10 PM
Repost from an original thread:
There is a lot of speculation in this thread. Readers might want to consider the following points when assessing the accuracy of the postings:
Industry Canada established a channel plan for DTV implementation. Existing television stations and planned channel allocations for additional analog television stations were given a DTV allocation. The DTV plan had to fit around existing analog stations and unused analog allocations. New DTV stations could not interfere with existing or planned analog stations or other planned DTV stations in Canada or the U.S. The resulting transition plan is characterized by low transmit power for the DTV stations. Some DTV stations are required to provide additional directional protection toward existing stations or new analog and digital assignments.
The station power in the Industry Canada plan is known as Effective Radiated Power (ERP) which is the amount of power radiated from the antenna. The ERP is not the power produced by the transmitter - it is the power after allowing for the antenna gain and transmission system losses. Increased antenna height on the tower usually requires a decrease in ERP. Higher UHF channels have higher allowed ERP because of higher signal losses over the air. For example, assume a 10 kW (kilowatt) transmitter. Assume the transmission system has a loss of just 3 dB resulting in 5 kW at the input to the antenna. The ERP will be determined by antenna gain, 3 dB gets back to 10 kW, 6 dB gets 20 kW, 9 dB gets 40 kW. You can see the pattern that every 3 dB gain doubles the ERP.
The Industry Canada plan makes certain assumptions about the type of antenna system used with the digital receiver. The relatively low maximum powers in the Industry Canada plan work with the assumed antenna system. Adding 3 dB of gain to the receiver antenna is almost the same as doubling the transmitted ERP. The Industry Canada plan does not assume rabbit ears in the basement as the antenna system.
As you get farther from the transmitter, the atmosphere plays a greater role in determining signal strength. Rain can reduce signal levels, weather fronts can cause reflections that cancel or enhance signals and storms can cause signal ducting that make very long distance reception possible. In the analog world, picture quality declines as the signal level drops. The picture gets snowy, goes to black and white and eventually breaks up when the receiver looses synchronization. In digital, picture quality stays constant even while the signal level drops until the digital decoder and error correction software cannot recover error free data. When this occurs, macroblocking and other problems appear and sound may be interrupted. As the signal level drops further, the frequency of visible and audible errors increases around the threshold of the receiver until the MPEG decoder cannot produce a useful picture.
So, in analog television, signal level determines the amount of snow in the picture. In digital, signal level determines the number of digital defects that appear due to uncorrected errors. If you have lots of signal then the digital decoder will rarely have problems recovering error free data and the picture will be perfect. If the average signal level is close to the receiver threshold then visible defects will be common as atmospheric conditions cause the signal to vary and occasionally drop below the receiver threshold generating uncorrectable errors. High signal levels also help the receiver overcome electrical noise from all types of electrical devices.
The U.S. DTV plan assumes the need for higher signal levels to work with less efficient antenna systems. As a result, the U.S. plan has DTV stations up to 1 Megawatt ERP such as WUTV Buffalo on channel 14. Stations located on higher terrain have lower ERP such as WIVB Buffalo with 960 kilowatts on channel 39. The Canadian DTV plan maximum level is less than 10% of the U.S. maximum so a good outdoor antenna is required. There is some indication that Industry Canada might review the plan to permit higher power for DTV stations.
I hope this is helpful to understanding DTV over the air transmission. If you want to get more information on the Industry Canada DTV plan then go to the web site. For more technical information on DTV you can check ATSC.org.
2005-06-07, 09:00 PM
The Toronto digitals are sharing transmission antennas on CN tower with analog signals at close to same frequency.This is why the power levels are low.To prevent interference to the analog tv signals and save the expense of installing new antennas on the CN Tower.If the analogs were turned off then power levels could increase.
2005-06-07, 09:06 PM
That's a good point, but it doesn't account for the low levels being decreed elsewhere, such as in Vancouver.
2005-06-08, 09:28 PM
Doesn't explain CKXT-2-Hamilton either :) (only at a paltry 4.1kW).
2005-06-09, 11:46 AM
Does it undergo compression/recompression in the process?
No it doesn't. At least in terms of "lossy" compression.
But, try as they might, the providers cannot ensure that the original signal is pristine and clear of degradation or artifacts by the time it has made its full journey to the home receiver.
Perhaps they cannot, but transmission errors would show up as data errors on the receiver. Most of these errors would be visible in the picture.
For a Vancouver subscriber the signal's round trip can account for a visual and audible difference when compared A-B with a direct OTA signal.
Not unless something is intentionally being done to the signal.
2005-06-09, 02:43 PM
No it doesn't. At least in terms of "lossy" compression.You mean other than the lossy MPEG1, MPEG2, and MPEG4? What is the unlossy compression system being used?Perhaps they cannot, but transmission errors would show up as data errors on the receiver. Most of these errors would be visible in the picture.Exactly! Bear in mind that I have seen the big difference between BEV HD and OTA HD with my own eyes, so what I'm about to say is based on my background in data transfer technology as I hope to explain my conjecture that OTA DTV is "cleaner" than most CATV/DBS systems (which are really just data networks).
In data transfer, packets actually are mostly labels with very little real "payload" in them, so if those labels get screwed up in just one packet it will be dropped (gone forever), which equals a tiny gap in the picture or sound. If the packet drop happens upstream in the network the error checking algorithms will rapidly cause it to be resent, but the problem is that real time TV data is serial like a freight train and does not like to be stopped so that a missing packet can be resent and take its place in a queue. Microseconds can make a big difference in streaming media.
If the packet drop happens near the receiver a possible visual disturbance might happen but with just one packet gone it would be faster than the human eye can detect, so nobody would notice. On the other hand, if the actual payload had a 1 and 0 accidentally transposed *ANYWHERE* in the stream, this is your basic "data corruption" and there is almost no possibility of error checking by the time it gets to your receiver. The result could be subtle enough that an image detail on the screen might be tainted as being carmen red when it should have been fire engine red. The result would be that the Ferrari on the screen might not look as sharp as on the original master recording.
Its my conjecture that while OTA DTV will suffer image quality problems like pixellation due to weird phenomena like multipathing or serious weather, there's almost nowhere in the point A to point B direct broadcast for data corruption to occur. This would make OTA DTV cleaner, and would explain much of why the visual difference to anyone not on certain Rogers HD CATV systems is easily appreciated.
With the low power levels being assigned by Industry Canada/CRTC there are unfortunately more opportunities for data corruption than from U.S. DTV stations under the exact same circumstances if it was possible to put them relatively side by side, such as around Sault Ste. Marie, Ontario.
Having said all this, I don't hold myself up as having all the answers, so I'll do some web reading and see what I come up with. Let me know if I'm going in the wrong direction here.
2005-06-09, 04:21 PM
What is the unlossy compression system being used?
I have no idea if there is any. I thought that maybe there might be data packet compression between some networking equipment, but that would affect any and all data traffic. Obviously, this wouldn't alter the data in any way.
Data errors are very visible in an MPEG2 bitstream. They will show up as green blocks on the screen. There is plenty of FEC in off-air ATSC bitstreams - much more than is required for a nice 'quiet' transport like cable. With this and reed-solomon coding, there's plenty of opportunity to accurately correct transmission errors - both for off-air, and for cable and satellite. Cable is using simple remodulators that demodulate the off-air 8VSB signal, and remodulate it to QAM. There is ZERO DATA LOSS in this procedure.
Like 57 said, most satellite companies are either using rate-shaping to reduce the bit rate of the channel, or are decoding, then re-encoding to a lower bitrate. In either case, there will be some reduction in image quality.
So, you cannot make a valid comparison between off-air and satellite, but you GENERALLY can between off-air and cable. I say generally because what exists today may not exist tomorrow. But right now, the MPEG2 transport stream on Rogers cable for ABC Detroit is bit-for-bit identical to receiving that same station off-air.
2005-06-10, 02:49 PM
So, you cannot make a valid comparison between off-air and satellite, but you GENERALLY can between off-air and cable. I say generally because what exists today may not exist tomorrow. But right now, the MPEG2 transport stream on Rogers cable for ABC Detroit is bit-for-bit identical to receiving that same station off-air.I think this now comes back to a point that seems to reflect today's situation across the country: if your CATV provider is serious about DTV, and especially HD, they will provide the highest quality that the standards and technologies offer. Unfortunately that is far from the case. In this area DBS in Canada is a big disappointment, so I've written off BEV and *C for high performance DTV. This leaves OTA, if you happen to be lucky enough to live in an area that has it.
2005-06-10, 07:59 PM
In response to Ryans ckxt-2,tor 1 is using uhf channel 15 which is also used in Buffalo for an analog station.Conflict would limit power and I bet it shares antenna with tor 1 analog.Stations do not want to spend for two high power transmitters.Crtc should mandate a timetable to have same coverage as analog signals.The USA has manadated this for the 4 major networks.If tor 1 Toronto transmitter had suffient power the Hamilton ones would not be even needed.The antenna sharing is what is limiting power in Canada.
2005-06-22, 07:25 PM
I was wondering if somebody can answer my question for me. I'm new to this OTA stuff and I was wondering why analog even exist? What I mean by that is anybody can get TV for free, I don't understand why the Can. gov't allows it. Is it by law that they have to broadcast these signals or is it just a matter of companies broadcasting their signals to other markets and anyone in a good area can pick off the signals? Any help you guys can provide would be much appreciated.