• DTV Antenna Help and a touch of SNR, BER, & MER



    If you go online to any number of TV sites (TVFool, antennaweb, or hdtvprimer), you will be told which antenna has better gain and/or which antenna should work in your area.

    Gain is not the entire picture when dealing with DTV signals. There are various means of interference and even if one antenna has lower gain, it could get a particular station, or stations, better than one with clearly better gain if it has better rejection of some particular type of interference causing problems with some of the stations in your area or, in particular, at your location.

    With the old analog signal, we could actually see various forms of interference, such as reflections, weak signals, co-channel or same-channel interference. With DTV, the TV simply loses lock and the sound/picture drops out. This is where some expensive tools come into play.

    I've picked up a Digiair Pro and, so far, I've only used it to tell me what the signal levels were of various types of bow tie antennas, not if the signals were good or bad prospects for viewing. The tests I was running were just for relative signal-level/gain information. If, as GroundUrMast says, I had my own test range (and tools to produce and test the DTV RF signals), then I'd be able to say more. I'm hoping that, with readers helping, we'll find better ways to test antennas so that weíll all end up with better information to help us get those lost stations back. We may also end up being able to give better advice to others.

    To better test an antenna, we'd need to be able to produce reflections, same-channel interference, co-channel interference, etc., in a manner identical to real life situations. It isn't going to happen. The Digiair Pro is a good, less expensive, DTV signal analyzer, and I can see some of the data needed to tell me that, even though the signal is strong, it is not going to be an easy watch due to signal-to-nose or bit error rate, etc. It is, however, not on the scale of a multi-thousand dollar test instrument. Solidsignal.com seems to have a rather good price on this instrument at less than $325.

    To take this to someone's house and help them out is going to be a labor of love, not one of monetary gain. There is no way someone could pay for all the time it could take to figure out why a particular station is not being received well even though it has good signal levels. That time and effort gets multiplied by the number of difficult stations the person wants to receive reliably.

    Those of us trying to test things and figure things out can only advise you so far. The proof is, unfortunately, going to be in what you buy and try to use. Your report telling us how it worked for you may help us better help the next person. But again, this may not be helpful to someone else in a completely different situation.

    Before cable and satellite TV, we could ask the person in the local TV service store what worked where, or we could depend on our neighbors who had this or that antenna and could tell us how to point it. There were people to ask what worked and what didn't and they lived in the same area, usually right next door to us. Today there are very few who still even want to use Over-The-Air (OTA) antennas, let alone know anything about them. So we band together in places like this and try our best to help each other out.

    This does help, but I don't live where you do, so I can't really tell what will work best for you. About the best any of us unfamiliar with your area can tell you is what looks like it should work in your area. But, that's not necessarily the same thing as what will work in your particular location.

    Let me give a quick personal example. At my location, TV Fool says I should receive about 10 stations easily with an indoor antenna. On the other hand Antenna Web says I will need to put an antenna at about 60 feet to receive any stations from Washington DC. So what's the truth? The truth is actually a little of both. I found a good location using an antenna with some gain placed about six feet off the ground and mounted inside my garage. I'm still testing antennas and locations even after three years because I havenít found that perfect location which will get signals from both Baltimore and Washington all year round without any drop outs. The antenna in the garage is clearly better than the two I have mounted higher in my attic and about 15 feet to the west, pointed between the two cities, but it is still not perfect. The unfortunate thing is that there may not be any perfect location for my antenna unless I get it above the tree line, which is presently about 50 feet.

    Once again the "key" is LOCATION just like in that cute comic used on the ANTENNAS page with Beetle upside down and hanging from the roof and the antenna in his hand. Proper pointing and location of a directional antenna are still two important keys to good reception; unfortunately they are not the only important keys.

    ************************************************** ***************************
    Here's a comic from http://www.hdtvprimer.com/ANTENNAS/comic.gif
    © King Features Syndicate.
    See the article on "Choosing a mounting site" by Ken Nist, KQ6QV
    ************************************************** ***************************



    SNR, BER, and MER:

    Letís go over the three other measurements which can give us information about the signals weíre interested in. First is Signal to Noise Ratio, seen as either S/N or SNR. Iíll use SNR to be consistent with the next two abbreviations. The second is Bit Error Rate, BER, and the last is the Modulation Error Ratio, MER.

    Iíll give references to all of these signal measurements/qualities so that you can read more if youíre interested. They are all much more involved than I will get into here. They have already been written about in more detail by others.


    Signal To Noise Ratio (SNR):
    The basics of this quality are rather easy to understand. This is how far above the noise the signal is. If it is high enough above the noise, the receiver will be able to decode it and use it. If not, we say itís either below the noise floor or lost in the noise. We can think of a crowded room with a friend standing next to us telling us something and not really being able to hear well enough because of all the other conversations going on. The noise floor - how much noise there is in the room - is so high and loud that the signal, the voice of our friend, is lost within the noise. Our friendís voice is simply not loud enough, not high enough above the noise floor, for us to hear well enough to understand the words being spoken. We canít decode the signal. There is much more to it, plus there are some rather interesting methods which can be used to enhance the signal such as using a narrower bandwidth, which would be like being able to tune our ears only to the voice of our friend. You can see how this would really help to hear your friend. We would end up hearing far less noise so the noise floor would be lower in just that limited vocal range than in the entire range we can normally hear. If the voice of our friend doesnít change in volume (signal level) it would end up being louder than the noise, higher than this new noise floor, and weíd be better able to hear, receive, and understand or decode, the message or signal.

    References:
    Signal-to-noise ratio
    Signal to Noise Ratio, SNR

    Bit Error Rate / Bit Error Ratio:
    This is also easy to understand. It is the number of bits which are in error compared to the total number of bits sent. If I send you six bits and three of them are wrong, there is a 3/6 = 0.5 BER. Think of it this way: if I say, "you are a most wonderful person," but you hear, "XXX are X most XXX person" you canít decode the message, you canít understand what Iím trying to say. If on the other hand your hear, "you are X most wonderful person" the BER is 1/6 = 0.17 and here we can understand the message, but even here if an important word or bit is the missing the message could be lost, as in, "XXX are a most wonderful person" or "you are a most XXX person." However, the lower the BER, the better the signal and the more probable the error can be corrected and the signal recovered.

    References:
    Bit error rate
    BER Bit Error Rate Tutorial and Definition
    and
    Bit Error Rate: Fundamental Concepts and Measurement Issues

    Modulation Error Ratio, MER:
    I consider the MER to be the most difficult to understand because it involves things only radio wave related. We have to have a radio wave and then a way to modulate it, or impart information onto the radio wave. The closest I can come to an image to try to explain this is to think of a piece of paper as the radio wave; the modulation is the words printed on the piece of paper. This isnít quite correct because the words donít change the paper as modulation changes the radio wave, but itís close enough. The different types of modulation could be the different languages one can use to write the message. If the words are misspelled, we have one type of modulation error, and if the ink is smeared so the word is unreadable, we have a different type of modulation error, and the constellation point(s) for this word, this part of the modulation, is not in the correct location. As far as I know, the MER is only for digital signals with digital modulation. Each different type of digital modulation has a different constellation diagram. Weíre only interested in DTV, and so the only constellation diagram we need is the one used to represent the DTV MER. With the proper tools, one can actually figure out, with the help of the constellation diagram, what type of interference is causing the problem(s). This feature is not within the features of the Digiair DTV Signal Analyzer and probably only available in those multi-thousand-dollar analyzers. Still, the Digiair does show us an MER and perhaps, with a bit more reading and understanding, Iíll learn differently.

    Relating the vector and the constellation point is the statement, "MER is a measure of how much difference there is. The vector difference between the ideal target symbol vector and the transmitted symbol vector. Expressed in dB." Modulation Error Ratio - Mer - For those who may remember their math, a two-dimensional vector has two qualities: the length/amplitude and the angle which completely describe the vector.

    While MER is mostly measured at the transmitter to inform the technicians there how healthy the transmitter is, as stated above, it can help the installer figure out some problems. In the dtvforum link above, bellotv gives a very good example of this in postings #10 and 19. He used the bad MER value and the good BER value to figure out that the problem was the masthead amp, which is the amp right up at the antenna. I think this comes mostly from experience, but now we know, too.

    References:
    Modulation error ratio
    Constellation diagram
    Modulation
    Modulation Error Ratio (MER) and Error Vector Magnitude (EVM)

    If you want to read even more, simply do searches on the topics and Iím sure youíll find more than enough to keep you busy for some time to come.
    The Bit_error_rate topic on Wiki has no references and a warning because of that, but in reading it I believe it is a good article and agrees with what Iíve read elsewhere.

    Safety First:
    I simply can not stress this enough. Almost every year I hear of someone that has died while installing an antenna. So please, PLEASE be very careful if youíre installing an antenna on a roof or other elevated place. Donít slip like Beetle did, and donít place it too close to power lines if there are above-ground lines. No line should, if it breaks, or the pole it is on breaks, be able to touch your antenna, and the antenna should not be able to fall onto any power line. When installing antennas, always think worst case scenario! Testing an antenna roof location is always safer done with two people if possible: one to hold, move, and rotate the antenna and the other to read the meter. Never try to walk on a roof while reading a meter or while youíre holding or pointing the antenna. Always watch where youíre walking and whatís around you and your feet. Think safety first, look where your feet will be going, then move there only if it is safe to do so.

    Good luck with your antenna buying and installation.
    Phil KE3FL

    All my blog articles are listed at: Karras' Corner
    or
    Karras' Corner Article Links on my KE3FL web site.

    Karras' Corner Article List

    Like what you've read? Register and leave a comment...
    Already registered here? Type away!
    Comments 4 Comments
    1. NZ1J's Avatar
      NZ1J -
      Thanks for these articles. They're very helpful. I've built some parabolas for DTV and I've had good luck with them. I've built a 4 bay antenna, but used folded dipoles. I will try bowties now. Also, I've done some experiments with Yagi's, but I've had terrible results. I know the bandwidth is small, but I'm testing only on the one analog channel that I can receive.

      I may purchase the digiair, but for now I'm trying to determine signal strength by analog picture quality. I have never found screen shots that show, for instance, how 3dB of signal improvement would change a snowy analog picture. I'd be interested to hear what you think about testing Yagi's with TV signals (some commercial TV antennas have do have a lot of parasitic directors) and about signal strength effects on analog picture quality.

      Thanks and 73
    1. Stuart Sweet's Avatar
      Stuart Sweet -
      Gain is important when you look at antennas but the other metrics are equally important. It's been pointed out in other articles that additional gain, especially through an amplifier, won't compensate for a signal that is so noisy that it's impossible to lock on. That's not not say a larger antenna isn't worth it, but with digital signals especially there comes a point where no amount of gain will help.
    1. VOS's Avatar
      VOS -
      Quote Originally Posted by SS@SolidSignal View Post
      Gain is important when you look at antennas but the other metrics are equally important. It's been pointed out in other articles that additional gain, especially through an amplifier, won't compensate for a signal that is so noisy that it's impossible to lock on. That's not not say a larger antenna isn't worth it, but with digital signals especially there comes a point where no amount of gain will help.
      My experience has been:
      1. The signal has to have a decent signal to noise ratio (SNR) at the location.
      2. Multipath will degrade reception and the SNR
      3. The choice of antenna can help with multipath, but not enough for a low SNR.
      4. Digital barely locks on to a 15 dB SNR. Above a constant 20 dB SNR has given good reception.
      5. Websites like TVfool can be helpful for channels, strengths and directions, but are only estimated levels and can't account for local effects like foliage, buildings, etc.
      6. The antenna and type is the best gain as it doesn't add to the noise.
      7. A preamp can't overcome lack of gain from the antenna, but can help the SNR to the TV, if the signal from the antenna has a good SNR.
      8. BER & MER seem to fall in place with a decent SNR.
    1. PhilK's Avatar
      PhilK -
      Sorry for this being so late.

      NZ1J, your use of a Yagi (Yagi-Uda) will work fine but only for one station, if that. Remember that most Yagi antennas are developed, in amateur circles, for bandwidths that, when compared to the 6MHz of a TV signal's bandwidth (BW), are very small. The entire 20 meter band is just 350 kHz, and the 6 meter band, very near old channel 2, is the one that gets close at 4 MHz.

      The antenna you want, if you're looking to cover a good portion of the TV bands is the log periodic antenna. One can think of the log periodic as a group of three element Yagi antennas one reflector, driven element, and one director, the director of one channel becomes the reflector for the next channel and the reflector of our first channel is the director of the channel below it.

      Also, the spacing of the directors and reflectors to the driven element and each other is important and there are equations for figuring it out, however, the more directors and reflectors the more narrow the BW of the antenna so while the gain goes up the BW goes down. Not exactly what's needed for a TV antenna.

      The bow tie antenna and a group of them, over come these problems, at least in theory, by being extremely broad BW antennas to begin with which makes, the angle, the distances, and the over-all length all very forgiving. My tests confirm this and while changing these things does effect the overall performance of the antenna it in no way improves the entire range of frequencies so that I can clearly say that this angle, length, or separation distance is the best for all TV channels. I only saw minor improvements in TV channel sections for any change made with corresponding decreases elsewhere.

      Here's the wiki article on the Log Periodic antenna:
      Log-periodic antenna - Wikipedia, the free encyclopedia

      And one on the Bow Tie antenna by a PhD Antenna Engineer:
      http://www.antenna-theory.com/antenn...and/bowtie.php

      I hope this help, definitely look up the log periodic antenna and give the bow tie a test to see what helps.