Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uwm.edu!rutgers!att!cbnews!military From: denbeste@spdcc.com (Steven Den Beste) Newsgroups: sci.military Subject: Re: "Identify-Friend-or-Foe" questions Message-ID: <12972@cbnews.ATT.COM> Date: 10 Jan 90 03:28:15 GMT References: <12566@cbnews.ATT.COM> <12883@cbnews.ATT.COM> Sender: military@cbnews.ATT.COM Organization: S.P. Dyer Computer Consulting, Cambridge MA Lines: 94 Approved: military@att.att.com From: denbeste@spdcc.com (Steven Den Beste) I've gone through this in private with someone, and apparently everyone has misunderstood my question, so I'd like to go through it in public. I am not contending that an enemy SAM could easily make itself look friendly (and therefore be ignored by the jet) by simple recording. In the book "The Threat: Inside the Soviet Military Machine" (much recommended, by the way) Andrew Cockburn describes how a U.S. SAM was tested against a drone in front of several congresspersons. The test was quite successful. Of course, what the congresspersons were NOT told was that the drone carried a radio transponder, and the missiles homed in on that. Needless to say, this made the success rate much higher than if the drone was trying to evade, or if the SAMs had to find it on their own.. >From an operational point of view, a jet's IFF-receiver is precisely such a transponder. Whenver it receives a challenge, it answers. If a SAM can figure out how to send a challenge, it can then home in on the jet's answer. It doesn't know what the answer means, and it doesn't care. At the lowest level, it is just a radio beacon towards which the missile can aim. The jet will, of course, also challenge the missile and it won't get the correct answer back. The jet will, therefore, correctly identify the missile as a foe, and try to evade, destroy or confuse the missile through a legion of interesting techniques. But if the missile can record a valid challenge, no matter from where it came and whatever it means and keep sending it, then the jet will keep answering and the missile can home in on the answer. Anything which prevents the jet from answering also risks having the jet fail to answer a missile which IS friendly. Thus long tailpipes or dropped flares or ECM radar fuzz have no important effect on this missile, since it is only homing in on the IFF answer, which it can locate even though it doesn't understand it and can't decipher it. I've come up with three ways the system could be made to work, and none of them is practical: 1. All friendly units have a time-of-day clock with a fast tick-rate (no more than a few seconds per tick) which are very closely synchronized. Part of a legal challenge contains the time-of-day. Any challenge which contains the wrong time-of-day is ignored. Thus, by the time the missile has taped the challenge and repeated it back, the challenge is already obsolete. Since the missile doesn't actually know the enciphering algorithm, it cannot synthesize its own non-obsolete challenges. Just how do you make sure that all the friendly time-of-day clocks are synchronized within a fraction of a second? I don't see how this could be done in practice. 2. A legal challenge contains the type of unit from which the challenge was received. When avionics on a jet receives a challenge, it makes an independent identification of the type of unit, and if the challenge makes no sense from that type of unit, then it is ignored. In other words, a missile broadcasting a jet-challenge must be an enemy. My private correspondent said that this might be possible by monitoring the kind of forward radar the unit was using, since this seems to vary substantially by kind of unit. It seems to me that all this means is that the enemy who designs this SAM must figure out how to broadcast a jet-like radar signature (even though it isn't actually using radar to find its target - remember, it is ONLY homing in on the IFF response). This becomes an arms-race, where the jet-building side tries to get smarter and smarter about identifying bogies and the SAM-building side tries to get cagier and cagier about looking like a friendly jet. 3. A sequence number is used theater-wide and is included as part of the challenge. Once any unit in the theater uses a challenge with a given sequence number, that sequence-number never gets used again by anyone. Perhaps this could be made to work if every individual IFF has its own range of use-once-and-never-use-again numbers. (Good thing numbers are cheap! We're going to need a lot of them!) In that case, seeing a message which uses a number that has been used before, or is in the jet's own number range, would be ignored. Again I think this is a logistics nightmare. It also requires all the IFF systems to contain enormous memories to store which sequence numbers they've seen before. So my point is this: As long as one of our jets has IFF onboard and operating, there should be a relatively unsubtle way that a SAM can home in on that IFF system so as to destroy said jet. The jet's IFF will correctly identify the SAM as a foe, but this is irrelevant. (Except that it means the pilot can try somehow to evade the thing.) I'm hoping that there is some other approach besides the ones I've dreamed up which overcome this problem. Would anyone care to take a crack at explaining it to me?