When we’re thinking about SIGINT, excessive power and bandwidth are not conducive to a low-profile survival radio strategy!
It’s often quite difficult to connect all the dots on the blog and channel. Hopefully this post can help! One side of it is avoiding detection when we’re thinking about signals intelligence (SIGINT). Every time a PTT is pushed, a timer begins. It’s only a matter of time before someone figures out who we are, where we are, and what we’re doing. In everyday life, this is no big deal. When SHTF, we should have a better understanding of physics. Although it isn’t explicitly said on the channel, reducing my electromagnetic footprint, is a big part of the program. Let me explain.
Many of us start out in the preparedness Community proclaiming our intention to buy a radio like the Yaesu FT-857, which is an extremely fine radio. In fact I’ve owned one of them in addition to an FT-897. Both are/were fine rigs! When operators get those radios out in the field, we often do so without thinking about the amount of output power we’re using. Generally that’s 100 Watts. This often happens because we’re still mixing up amateur radio for DX versus amateur radio for preparedness. The seasoned operators understand the concept of using only enough power to make the contact, but we’ll come back to this shortly.
Other operators start off with V/UHF radios for “squad” level voice communications. Many of the latest V/UHF models have more than 5 watts output for greater range. In a way that higher output power is pure marketing, but we’ll come back to that in another post. Like our Yaesu FT-857 manpack brethren, there is this mistaken believe that the radio has one setting for output power. Usually that is the “high” setting, regardless of the antenna connected to the radio or distance to the receiving station. The fear is, we must use as much power as available, to get the signal through. This is not wrong, but physics can offer some alternatives.
Here’s the thing. Voice Communications on FM, DMR, DSTAR, SSB even packet and APRS are extremely wide! Wide bandwidth voice or data modes require much more power to be effective than a narrow mode like JS8. The wider the bandwidth, the more power it takes to get that signal out into the world! The narrower the signals bandwidth, the further that signal will go, on the same amount of output power. Because of that narrow bandwidth, we can reduce our output power, without losing much effectiveness. At the end of the day, survival radio is all about “Reliable & Effective Communications”.
You often see me talking about radio efficiency on the channel. This is often connected to the “how we power our radios” discussion. It’s also related to the amount of power we need to get our signals out. Wide bandwidth signals like voice modes, require an awful lot of battery power to push those signals out into the world. Think about it like the difference in shooting a shotgun slug versus 5.56 NATO or even a .22 out a couple of hundred yards or meters. Think about the energy it takes to get that slug down range versus the smaller round. They’ll definitely both make it down range, but one will make it down range with much greater efficiency (less energy) than the other. The shotgun shell is like our wide bandwidth voice modes and the 5.56 is like JS8. Another good analogy is spot light versus the flood light. A spot light can reach farther, but effectively focuses on a single point. A flood light spreads its light over a wide area but doesn’t go very far. Does this make sense?
OK, so what does all this have to do with SIGINT? Lets look at the next image.
Each of the narrow squiggly lines in the above image is a signal. Powerful signals are easy to see, as they are much brighter in the waterfall. From midway in the waterfall to the right side of the waterfall, there are strong signals. These signals are easy to spot, for the SIGINT operator trying to detect or listen in on your communications. Now look at the waterfall again. This time midway in the waterfall and to the left. There are still signals present in the waterfall, but they are using much less power. Some of them are even using less bandwidth than others. There are also some which are almost impossible to see with your naked eye, but they are there! Those difficult to see signals often get passed over by a human, as they are indistinguishable from background noise with the naked eye. If we use lower power, it will be more difficult to see our signal. Using lower power also makes it more difficult for a human to hear our signals while receiving. Combine that with narrow bandwidth data modes, we are effectively hiding in the noise. Civilian SIGINT operators rarely have the level of sophisticated communications gear, to dig a -23db signal out of the noise. Add in a high noise floor, and that station becomes quite blind electromagnetically speaking.
You see, because of the operator fatigue envolved with scanning the waterfall for our signals manually, civilian SIGINT operators are focusing on strong, wide signals. Signals which are easy to see or hear in the waterfall. Government, Military and NGOs most certainly automate the signal detection process. For them it is better to fly low and avoid the radar. Even so, from the moment our signal is detected, there is a chain of command and logistics process initiated, before any action on our signal will take place. How much time is our best guess.
I understand this is an advanced survival radio topic and more abstract than most. Apologies for that, but it is what it is.
- Understand the electromagnetic footprint of your chosen mode.
- Use as little power as possible to reduce that footprint.
- Make up for using lower power by using narrow bandwidth data when possible.
- Take advantage of the increased operating time and greater efficiency of narrow bandwidth comms.
- Take advantage of the increased security of low power, narrow bandwidth comms.
- Understand how low power and narrow bandwidth can also reduce your comms loadout.
- Incorporate modes like JS8Call, VarAC 500hz, VaraHF 500hz, into your communications plan.
- Utilize NVIS to further mask your RF footprint from civilian eavesdroppers.
Buy me a rootbeer: https://paypal.me/oh8stn/2usd
Excellent, thought-provoking content, as we’ve come to expect from Julian.
Thank you 🙏
Chapter 2 in the “Work vs Play” series. Good stuff. 73
Julian – very thoughtful here, thank you. To further the concept of efficiency, maybe we can think another concept like time on the air. We might debate whether wider bandwidth and more power might be better w.r.t. detection/location. Here’s a thought — your plot there is describing power spectral density (PSD), a measure of power per Hz.
If you are hearing my transmission of 10W on a VaraHF 500Hz Winlink call, you’ll see it (500Hz wide signal) as some brightness depending on the path loss between us. If I chose to send the same email at full BW (let’s say ~2500Hz) at 50W, you’d see the same brightness of a waterfall because it was the same PSD (five times the BW and five times the power)… but you’d see it for 5x less time on the air.
If you agree that when using a spectrum scope with a resolution bandwidth (the size of the frequency “bin”) in the Hz range, then the only way to reduce the brightness is to reduce the PSD. Running narrow BW makes this higher, not lower, for the same power.
Final thought — for digital comms where there is a fixed length payload to move allows:
• the exchange to happen faster
• frees up the medium (channel) for others
• gets you off the air faster to avoid detection if being done real-time
• might offer a higher efficiency operating point for your transmitter — less energy consumed for the message
To my signals-and-systems friends — Are there un-encrypted spreading techniques like CDMA that we can employ? The norm seems to be to improve received SNR by reducing the channel bandwidth… but with coding techniques we can burn power and time on the air by sending many more bits than we have actual information to send, and spread the resulting coded message over a very large BW — hiding in the noise, and relying on de-spreading to recover. I’ll look into this.
Excellent points and lesson.
My theory is we not only reduce bandwidth but reduce output power. Then take advantage of data modes which can decode at very lower signal to noise ratio, like JS8 or VaraHF. We’ve tested this many times with JS8 and VarAC. The operator on the transmitter side reduces output power, while reducing bandwidth. This is easier to see with js8. As power is reduced, that signal in the waterfall fades out of view from from the receiving stations waterfall, while is still being perfectly readable by the software. VaraHF wasn’t probably the best example. JS8call certainly is with its ability to decode far below the noise floor.
I agree, we spend much more time on air as the narrow bandwidth data signal means longer time to send the message. Keep in mind we’re using the narrow bandwidth because we either don’t have available or don’t want to use higher power. So the question is really which one is more efficient. Sending slow data over a longer period of time, or blasting out that same amount of data in a shorter period of time but with higher power!? If we consider the efficiency of the radio. An Icom IC705, TX500, KX2 are far more efficient than an 891 or 7300 we would use to send out higher power. This is true even at the same output power eg 10 watts. The overhead of the QRO radios is quite poor.
I wonder if it’s a valid point when considering the higher output power used, the more equipment we actually need for the station in the field. Bigger batteries, bigger solar panels, more capable charge controllers, higher current consumption from more capable rigs. All of this just so we can send the messages out faster.
It’s a good discussion.
What sort of Civilian SIGINT comes to mind? US Amateur Radio bands are not limited in surveillance by anyone.
Use your imagination. The scenario is based on post SHTF, where it would be logical to minimize ones RF footprint.
A new YouTube video for your consideration :
Tracking airplanes with an SDR? I watched this video previously. Just don’t see the SIGINT connection from a preparedness perspective.
It was otherwise a good video with one exception. Focusing on commercial aviation rarely has any tactical value in peace time. In war time, transponders are disabled. If their focus was on a more pragmatic use of SIGINT, I’d be totally on board.
Otherwise it was a good video.
thanks for sharing.
Spread Spectrum Techniques was quite the subject of communications during the Cold War. For those unknowing to this it is burying an encrypted message in noise.Without the proper equipment the received signal is pure noise.
To communicate you need a certain amount of energy per bit at the receiver.
There are to ways to establish this, firstly up your radiated power at the transmitter, secondly focus the power you have by using directional antennas that give effective gain. Of the two the second option is preferable for covert operation. At HF directional antennas are not exactly small thus this becomes a problem. Some talk of NVIS operation, however it only works against ground mounted seekers if they are airborn NVIS is of less use than a vertical dipole. The downside of antennas of any physical size is that they have setup and tear down times which hold you in a location way longer than sending a message. The speed of response by a seeker can be in the order of two minuits or less which means you have to consider not just physical agility but agility in transmission scheduals and frequencies. One of the down sides of many digital systems is synchronisation, it acts as a fingerprint thus events happen on an edge and a seeker can use this to their advantage. One way to look at agility is speed of movment, another is distance of movment whilst they might appear the inverse of each other in physical movment, the same is not true for frequency movment where MHz jumps are as fast as 100Hz jumps in modern radio systems with Digital VFO’s. Unfortunayely due to licencing requirments licenced Amateurs have to stay within archaic regulations which make the use of even basic Spread Spectrum be it direct or hopping not alowed also have a think about how you might put in encryption between the modem and the call handler, that is beyween say JS8 and Call. Most amateur software in no way makes this easy, so you are stuck with sending plaintext. Some might say well just type in an encrypted message, unfortunately as with morse code amateur software tends to use frequently used standard language lettets as shorter codes for an easy form of compresion so encrypted text typed in would be inefficient. Another trick that can be tried goes back to WWII and the “battle of the beams” whilst it is very difficult to make an antenna of very narow beam it is a lot lot easier to use two antennas, and just overlap them. As the antennas move further appart that overlap becomes a small area like the Direction Finding Cocked hat”. Whilst that does not immediately appear of use, consider that you use a common modem and very low power output modulated signal, that you then feed into two double balanced mixers with their LO port driven by a genuine “Truly Random Generator” the result is two direct sequence spread spectrum signals, which can only be decoded in the cocked hat. You can further this idea with “Multiple Input Multiple Output”(MIMO) systems which is the way some go for confidentiality these days. Sadly in the past Amateur experimentation was behind covert communications, these days the regulations we work by stop us moving things forward. However there are ways to get around the rules that restrict amateurs on of which is to experiment on dummy loads. Another is to gain excemption by using other licences where encryption eyc is alowed.. At the end of the day doing experimentation in the microwave ISM bands in your back garden is not going to get you into trouble in most places if at some future time it needs to be scaled up, some ISM is adjacent to HAM bands where high power equipment will be available and capable of working in a broader band than just the HAM band.