Return to SAGE - Historical Notes

Jamming comments

from John Wright - June 23, 2015

1 ECM and ECCM are old terms. See: https://en.wikipedia.org/wiki/Electronic_warfare

Table of Contents
     - Sage Background
     - The Jam vs CounterJam contest

Sage Background
I found the webpage through a group discussion about the SAGE system and its ability to deal with jamming in a World War III situation that has never happened (thankfully), if and when it was ever attacked by hostile bombers, presumably Russian. We played war games all the time, day after day, day after day. SAGE stands for "Semi-Automatic Ground Environment."

[For example: The SAGE building at McGuire Air Force Base, circa 1958. On the far left are cooling towers for the generators located in the (low) middle building. The "cube" has four floors, with air conditioning and wiring on the ground, the computers on the second floor, offices on the third and the combat center on top.]

The SAGE intercept direction center known Duluth Air Defense Sector (DUADS) was only one of a whole host of defensive intercept fire control centers originally designated, but it began covering a much larger area in 1963 when it merged with Sault Ste. Marie Sector (SSMADS, to the east) and another sector to the west.

An old map
where I added red boundary lines to give you a rough idea of the area that we were controlling. What we did is like the setting of the movie "War Games," and I'll bet you saw it. The plot of the movie was preposterous, and the ghost track episode that I described in my letter was the only thing I ever witnessed that might be called an incident. Granted, the ghost was was coming straight toward our blockhouse, but we were on it immediately. Also, no one in that blockhouse believed that a single bomber would attack the US in any effective way; it wouldn't have survived long enough to reach the US/Canadian border.

The two main war rooms were Surveillance, where the radar input data were assembled into a map (i.e., battle intelligence) and Weapons, where interceptor missiles and manned fighter planes were directed to the targets on that map. There was also a small room where track identification was performed, where an unknown would stand out. We were all classified as "weapons controllers," and the dais platforms in those two rooms were no-lone zones (two officer rule), because our Bomarc missiles were tipped with tactical nukes. The surveillance dais had three commissioned officers:

RICMO - Radar Inputs and Countermeasures Officer (electronic warfare)
TO - Tracking Officer (in charge of the tracking teams).
ASO - Air Surveillance Officer (in charge of the entire Surveillance Room)

I served both as RICMO and (later) as ASO. More later. I have to go to a meeting.


You really ought to elaborate a little on the ghost track because if you don't, people are going to think it was some kind of flying saucer/UFO thing! It had a perfectly rational explanation, and the Staff Sgt. RICMT (radar inputs and countermeasures technician) got it right from the beginning (I really credit the man for clear thinking): The long range search radars scanned clockwise, just like the weather radars you see on most local TV weather channels. The new radar was painting an inbound airliner flying northwest to southeast (it eventually passed almost directly over the radar site). Our search radar painted that real object the way it was supposed to, with the antenna pointing to the northwest. But there was a metal building almost due north of the radar, and the FPS-27's transmitted power was so high that radar energy scattering from that building (i.e., by reflection) also bounced off the same plane and returned to the radar by the scattering path, except that the second path only happened with the radar antenna pointing north (i.e., toward the metal building). This resulted in a ghost signal in the northward direction. Because pulse travel time was virtually the same by either path, the northward ghost track painted at the same distance from the radar site, and it certainly looked real, coming right at the radar site! The fix involved removing the building.

What I told you in that first letter actually happened, and it stands out in my memory as a real life (though perhaps lackluster) version of the kind of thing the movie "War Games" tried to portray. In War Games the war room people couldn't discern between what was real and what was not, and the computer could do unacceptable things without human intervention. We could always intervene, even if it meant shutting something down (After the ghost track incident, the new FPS-27 was shut down and the old FPS-20 was turned back on). The ghost track incident shows that real and unreal are sometimes possible to confuse. When Pearl Harbor was attacked, a Major at the radar site didn't believe all those incoming radar returns were real, so he didn't act on the information. He could have saved the day. I think he ended up with a court martial.

Reiterating the one thing that still dominates my memories of the SAGE system: [One night shift, when I read the hot file, I noticed that a new FPS-27 had come on-line that day (I think it was at Armstrong, pretty much due north of DUADS). We had settled in for the evening, when the track identification line buzzed: "What's that uncorrelated data doing up there?" The TO had just prodded the responsible technician, and the active tracking began as the identification officer spoke. "There isn't supposed to be anything up there." We had an unknown. I got out of the ASO chair and moved to the RICMO position. The technician said it before I did: "I think it's some kind of ghost track. See this track coming down from the northwest on the trans-Canada airway? It has the same distance outward from the new radar." I replied, "Yeah, I think you are right." He moved to the RIM monitor (I didn't have to tell him to) while I called the radar site and requested that they put the FPS-26 on the northward track. The 26 found nothing, and the RICMT affirmed that the new search radar was the only one painting the unidentified. I called the SD and told him "It's not a real track, Sir. That's a brand new radar on line, and it's probably a reflection from something on the ground, nearby (later, it turned out to be a metal hut north of the new radar)." He replied, "You better be right, Wright!" But the thing went all the way to NORAD, and two interceptors went up there to confirm empty sky. I don't think the RICMO position ever needed a commissioned officer.]

The FPS-26 was a powerful height finder radar, one with some serious antijam capabilities. SD (actually SWD) was the Senior Weapons Director, the one in charge of the whole weapons room. Most of us referred to the SWD as "Senior Director." The WD (Weapons Director) had a lot to do with targeting decisions.

More later on my views of antijamming.

The Jam vs CounterJam contest
The special battle of electronic countermeasures (ECM, jamming) versus electronic counter-countermeasures (ECCM, antijamming)1 is mostly played out in the domain of time and radio frequency rather than of time and space, as in conventional warfare, and that is the reason why ECM and ECCM operators need reference to panoramic receivers that display signals on a time versus frequency display. ECM was a very real threat even when the SAGE system began, yet its best long range radar, the AN/FPS-20 had no ECCM capabilities. It would have been blinded during a real attack. In my opinion, the Russians would have trashed us. Perceiving this problem, an attempt was made to correct the deficiency by installing a variety of ECCM capable radars which operated at diverse radar frequencies; this concept came to be called frequency diversity (FD).

There was yet another problem, though it appeared to me not to be apparent to a lot of people involved with that system: tactics of the air battle between the interceptors and the incoming bombers were worked out at the SAGE direction center, whereas the electronic warfare battle happened one echelon down, at the radar installations. The two activities were well separated and did not have especially strong communications links, which - I think - was not a good idea. The ECCM buttons and panoramic video displays were located at the latter sites.

I was in the first group of students that went through ECCM training at Keesler AFB, where I learned the details of the various FD radar systems (see further down), and when I finished that course I was sent to the Sault Ste. Marie (SSMADS) direction center, to be a radar inputs and countermeasures officer (RICMO), not to one of the radar sites (where I could have used that training). Best advice to a RICMO was “Let the people at the radar sites handle the ECCM; stay out of their way because you can’t see the panoramic displays.” The RICMO could relay assessments of the jamming situation and its severity to the weapons director (WD); otherwise, there was not much he could do. Given the circumstances, I believed that a good NCO could replace the RICMO. Was there anything else the RICMO could do?

SSMADS was shut down and we were sent to DUADS (or at least most of us). In my case that other role happened at Duluth. I replaced a lost air surveillance officer (ASO) no less than four times (one was an actual fatality, when our ASO died in a T-33 crash). I spent more time being temporary ASO than RICMO (minimum rank for an ASO was captain, and I was still a first lieutenant). In desperation the senior weapons director (SWD) wanted to make me permanent ASO, which was not only denied by 30th Air Division but also used as an excuse to move me to 30th, to be its ECCM operations officer!

Just before I left DUADS some notices about an upcoming passive tracking capability reached us. Now that made sense, an ECCM capability that would work at the direction center end. It had come up in the Keesler training program. I wish I could tell you what became of that at DUADS, but I finished my active duty tour before anything happened. I don’t even know if actually became a system asset, but it does illustrate one kind of ECCM (and I will take up some others further down).

Passive tracking is a direction finding method that uses a jammer’s transmitted signal to find and track the jammer’s position. Turning on a jammer might blind an ordinary radar set by overloading its receiver, but the jamming signal can also be used as a beacon. Electronic warfare is not a one way street! This should call to mind missiles that can home on jam and anti radar missiles that home on a radar’s signal (as in the Wild Weasel attacks we used against Russian surface to air missiles in North Vietnam). ECM is a mixed blessing, and I think realization of that is what brought stealth into existence. There were some intermediate ideas, such as minimum skin paint coverage - holding the jamming signal to just enough (but the FD radars offered switch selectable receiver channels - ECCM or normal, etc. - and some receiver modes might reveal even minimal jamming signals).


Passive Tracking I

Passive Tracking II
Figures 1 and 2 (left) illustrate passive tracking.
    At the radar it was possible to measure received signal strength versus antenna azimuth (measured in degrees clockwise from North through East, South and West, then back to North). This created an azimuth versus amplitude display (AVA), where sudden spikes (increased signal strength) occur when the antenna’s beam passes through a jammer’s azimuth.
    The azimuth data at the radar sites would be digitized and sent to the SAGE direction center; at the direction center the computer would determine where the azimuth lines crossed.
    In Figure 1 the two red squares are two jamming aircraft and the radar sites sending AVA data are F (Finland MN) and C(Calumet MI). The AVA lines, however, have three crossings; the one labelled “g” is a ghost track - not real.
    Figure 2 shows a different situation where two jamming tracks and two radars lead to four crossings, thus, two of the crossings are ghosts. Things get really complicated if 25 or more enemy bombers are heading generally south with their jammers going full blast. That’s a more realistic WWIII situation.

Could it have been handled? You have to weed out the ghost tracks. I think the laptop that I’m typing on right now could be programmed to perform that task with great finesse, and these personal computers we take for granted today are far more powerful than the old SAGE computer! All aircraft fly at a characteristic cruise speed, but those ghost crossings behave differently - speeding up, slowing down, going steady at unexpected speeds. Yes, they could have been weeded out.

For all the money poured into SAGE, it seemed to stay several steps behind where it needed to be. That was because the Pentagon people and Robert McNamara believed that mutually assured destruction was our only true line of defense. I believe it got worse after I left.

So lets go on to those FD radars and a few things about them. First of all,
the background of the FD radar concept (in the historical sense) needs to be examined:

An even broader historical account that includes
pre-frequency diversity thinking and technology is found here:

And what came out of all of that testing and system development? Could those radars really deal with a heavy ECM environment. Some of the concepts were not actually compatible with the SAGE system even though they worked well if you were just viewing the radar video. The usable results were specific long range radar systems with ECCM capabilities
such as the AN/FPS-24 long range search radar:

The AN/FPS-35 long range search radar:

The AN/FPS-27 long range search radar:

The AN/FPS-26 heightfinder:

Although it pre-dated the above FD radars, the
L-band AN/FPS-7 long range search radar became part of the frequency diversity of the system:

The venerable AN/FPS-20 did have an antijam upgrade, i.e., to the AN/FPS-64:

An historical tabulation of radars types (over the years) is here:

So you had VHF radars (the FPS-24), UHF radars (the FPS-35), L-band radars (the FPS-7 and FPS-64 modifications of the old FPS-20; the FPS-7 could change frequency, though not rapidly), and microwave radars the (FPS-27). An enemy plane flying through the SAGE system would likely have double or even triple coverage from adjacent search radar sites, so if one intended to hide beneath jamming, it necessitated adding more jamming equipment to the bomber, thus a payload weight penalty. Basically, FD alone was not a bad idea.

A measure to counter ECM was called a fix. Generally, the fixes in the transmitter were the most effective, followed by those in the antenna system, followed by receiver fixes, and finally by those involving video processing.

TRANSMITTER FIXES

Of the transmitter fixes, the best was frequency agility:


The last FPS-35
That’s the ability to change frequency rapidly, to get out from under a jamming signal. The FPS-35 could change frequency on a pulse to pulse basis, but that disrupted the radar’s ability to eliminate ground clutter from relatively nearby signal reflections (the clutter eliminator was called “moving target indication,” MTI, and all of the radars normally operated with MTI in place). The FPS-35 could also suddenly change frequency just as it started painting a jamming aircraft, which might fool the ECM operator into thinking he was jamming the radar’s frequency. But frequency agilty was countered by broadband (barrage) jammers of the type known as carcinotrons -

Of course, the jammer’s power was spread out over a much wider frequency range, which only degraded the radar’s performance. Narrow band spot jamming is what blinds a radar set, which is what would have happened to the FPS-20.

Another transmitter fix was pulse compression:

In this technique the radar transmits a longer than usual pulse, say 20 microseconds long, and during that transmission the frequency is varied (chirped). If nothing were done the longer pulse would degrade target resolution, but the received signal was passed through a special electronic filter that compressed the pulse to, say, 5 microseconds, restoring resolution. It was as if the radar pulse was transmitted at much higher power, enough to cause dielectric breakdown in the waveguide, so it was a burn-through-the-jamming fix. It was not a selected fix; pulse compression was in operation at all times.

There were other transmitter fixes like staggered pulse recurrence frequency, which defeated some pulse-type jammers and eliminated interference from other (friendly) radars.

ANTENNA FIXES


Antenna Sidelobes
An example that comes to mind is the omni-directional antenna of those FD radars. Its vertical pattern and receiver gain were set to match the gain of the main sidelobes (lesser beams on both sides of the main beam - near 30o and 330o degrees, as shown in the diagram to the left). If pulses that appeared in the omni-directional channel matched the amplitude of those in the main receiver channel, then it indicated that a jammer was trying to deceive the radar (i.e., by creating ghost tracts in a different direction, hoping to divert interceptors to empty sky. Signals from the omni-directional and regular receiver channels are simply merged in a way that cancels the pulses that enter the side lobes, very effectively removing the deception jamming.

RECEIVER FIXES

SAGE radars had to operate at a constant false alarm rate (CFAR) to avoid overloading the SAGE computer. CFAR and some resistance to jamming was gained by Dicke and Logarithmic-type receivers, though not what I want to call spectacular. The Dicke method involved wide band amplification, followed by hard limiting, followed by narrow band amplification. Logarithmic receivers simply process signal amplitude logarithmically.

VIDEO FIXES

These eliminated some types of pulse jamming and aided when chaff drops blossomed. But one night during a SAC/NORAD Big Blast exercise a chaff drop was caught in the jet stream, north of the Canadian border, and the MTI of several radars along with clutter mapping couldn’t cope with it. It very nearly shut us down! After that I realized that weather conditions during a Russian chaff drop could mean the difference between a general thrashing and a total disaster. Never doubt that inattention to the electronic warfare situation can be devastating.


Two questions are worth asking:

If WWIII had actually happened in 1962 (the Cuban Crisis), what would have been the result?

The Russian bombers would have used a strike and roll-back approach, nuking radars and direction centers and fighter bases on their way to the other strategic targets. We would have shot a significant number of them down (and probably some of our own interceptors, too!), but we would have only blunted the blows. I very seriously doubt that I could have survived it. The cost in military and civilian lives would have been staggering. I’m glad WWIII didn’t happen on my watch.

Now that we’ve been through glasnost and perestroika and the Berlin wall coming down, can we feel safe from nuclear annhilation?

Personally, I think the prospect is worse now than it ever was.

But that’s just my own opinion. What do you think?

_______________________