Wednesday, March 27, 2013

Contemplating Privacy and Radio Communications II

Let's say you and I were out in a crowd - a party or something - and you wanted to tell me something that you didn't want other people to hear.  How would you handle that?  I'll bet you'd lower your voice and get closer to me, whispering it to me.  The radio equivalent of this is to lower your transmitter power.  You wouldn't want to stand on the tallest table on the other end of the room and use a megaphone to tell me, while scolding the other people for hearing it.  That's the equivalent of legislation.

The obvious question is how much transmitter power do you need to get your message to the other guy?  That depends mostly on the modulation type you're using, the distance, and the frequency.  If you want to be analytic and calculate it, you want to calculate what systems engineers call a "link budget" which centers on the "path loss" from transmitter to receiver.  You can find examples and background with a few minutes in your favorite search engine.  This one seems like a pretty decent online path loss calculator.  You enter a transmitter power, frequency, distance, and antenna gain, and get the power at the receiving station.

One of the things you need to to know to use these solvers is how strong the signal needs to be in order to be usable. In professional radio, they assume that a 12 dB signal to noise ratio (SNR) is good for voice work. I find that's pretty accurate for people with a lot of radio experience, they can often work with lower signals (they sound noisier), but inexperienced people might need a stronger signal.

So let's say I need to get a confidential message to a friend a few miles away.  We both have a ham license and a dual band HT (Handy Talkie - I think that was Motorola's trade name for a grown up walkie talkie).  Like most of these VHF/UHF radios, they use narrowband FM voice.  At VHF, these radios are usually very sensitive; they can deliver 12 dB Signal to Noise Ratio at 0.2 microvolt or about -120 dBm and sometimes even less.  If my friend is 2 miles away, the path loss using that link above is 86 dB.  That means that since -120 dBm is all I need to create at the receiver I can transmit at a very low power of -35 dBm.  I can use the lowest power my HT will put out and then attenuate it even more.  Ordinarily, they put out one watt (+30 dBm) or so on their lowest setting.    
I'm afraid that's going to read like gobbledygook to many.   A dB is a ratio of two numbers, 10*log (ratio).  A dBm is a ratio to a specific power: 1 milliWatt, .001W, in 50 ohms.  To be honest, more than half the engineers I work with use these terms improperly.  When I say "one watt (+30 dBm)" I'm specifying 30 dB > 1 mW: a ratio of 1000 times 1/1000 of a watt.  -120 dBm is a very small power: 10 to the minus twelfth power times 1 mW.  Once you get used to it, it's much easier to use than any other way of keeping track of many problems in radio work. 

I don't want to spend too much time talking about exactly how much power to use and how to get it because there's an amazingly big problem lurking just under the surface.  Let's say I get a power attenuator and reduce my transmitter power a million times (60 dB) until I'm putting just enough signal into my friend's receiver.  Furthermore, there's no other local noise, no big obstacles between us and no other things that could make me need more power are present.  2 miles away, my signal is essentially undetectable unless someone is tuned to exactly the right frequency at exactly the right time.  People 2 miles farther away are less likely to hear it since it has dropped another 6 dB.  Just what I want. The problem is that at my transmitter, I'm still putting out a signal that's pretty strong to a receiver.  Think about it: that signal is 86 dB (the path loss) stronger than it has to be for that receiver to hear it, and that's amazingly strong to the receiver!  If someone is trying to find me,  they just need to be within a few hundred yards of me.  This is sometimes called the Near/Far problem.  To make a signal useful far away, it becomes extremely strong to nearby receivers. 

If I use encryption, even if they heard the signal they wouldn't be able to use the intelligence it contains.  But maybe they don't really care what I'm saying.  Anyone transmitting in this town is assumed to be a terrorist and gets a SWAT raid - pick your scenario.  Plus, for a simple system like FM voice, all they need to do is generate a strong signal on frequency at the receiver's end, stronger than the expected signal, and they "capture" the receiver, making it impossible for the receiver to hear what they want.  

There are ways to try to get around their electronic countermeasures (counter-countermeasures).  Perhaps you change frequency regularly.  Perhaps instead of using a popular amateur frequency like 146 MHz, you use one that hardly anyone ever uses, like 1220 MHz or 10 GHz.  A narrowband, low power signal at 10GHz can be very hard to find, for a variety of reasons.  Whatever frequency you use, turn off your radio when you're not using it.  99.9999% of modern radios contain a "Local Oscillator" or LO (local means local to the radio) and these can be detected from some distance away.  There used to be a TV ratings agency that instead of calling and asking what you were watching, or putting a monitor box on your TV, instead would drive around neighborhoods watching the LO signals and determining what channels the TVs were tuned to. 

And if something big depends on this communications link, try it out while times are good and a transmitter doesn't get a Hellfire through your window. 

Much of this is going to go over the heads of people who aren't familiar with radio.  I'm aiming this at hams, GMRS or FRS users, and others who have played with radio but never thought about things in this way.  I apologize if it's too much geek speak.  And tell me if nothing makes sense.

More to come...



10 comments:

  1. I have that radio pictured.....great little rig.

    ReplyDelete
    Replies
    1. Yeah, they're a lot of fun. With a decent wideband antenna, they're surprisingly good at SW, too.

      I was going to post a pic of one of those Chinese.com dual band HTs, but I have no idea if any are good or bad, and I'd hate for someone to buy it based on my picture.

      Yaesu radios I can vouch for.

      Delete
    2. I love them. I have two handhelds, an FT-60, and an FT-270. I also have a mobile: FT-7900.
      I am currently studying for my General, and I will invest in another rig.

      Delete
    3. FWIW, I work with a couple of hams and they both recommend one of the ~$50 Baofeng models as a good starter unit.

      Delete
  2. Terrific info, thanks for posting. Looking forward to your take on digital modes and other options for more secure comms for SHTF.

    ReplyDelete
  3. Graybeard,

    Great food for thought. I need to beef back up on my RF theory, it's pretty stale.

    A question: Do you know of any available resources to build freq hopping and encrypted radios available to the public. It might be a good project to get some of your engineers working on, perhaps using this as a foundation: http://greatscottgadgets.com/hackrf/
    All you need is a laptop and the Hackrf USB radio, throw on a GUI that makes signal strength and frequency selection easy, encryption and frequency hop loads, and you've got a military grade radio.

    Thanks!

    Alan Turing

    ReplyDelete
    Replies
    1. You're getting ahead of me. I'm working on it, though.

      Delete
  4. Besides dropping the power level, a small (3 or 4 element) directional antenna can also help quite a bit.

    VHF elements are a bit long and unwieldy (38" @ 2 meters), but at UHF they get quite manageable.

    And by making the elements for a small Yagi out of a tape measure, you can just fold it all up and stuff it in a convenient pocket.

    ReplyDelete
    Replies
    1. The strength of the directional antenna isn't in the forward gain, it's in reducing radiation in unwanted directions. A 3-4 element antenna is only "a few" dB. If it can knock the sides down by 10 or 15, all the better.

      Delete
    2. And that was my point.

      Focus the energy where you want it to go, and you can use less power to complete the link, while greatly reducing unwanted pickup.

      Didn't come out as well as I wanted it to the first time, as I'm a bit under the weather, and my thinking is a bit fuzzy!

      Delete