06 April 2012


man-in-the-middle attack occurs when a third is located between two principals, whether individuals or some type of equipment, and deceives them into believing that each of them he is the other principal. It acts as an interface for modifying the flow of information exchanged. The only solution to this type of attack is the encryption of communications using a protocol that involves a key exchange through a secure channel, so that even if an attacker could intervene communication would be unable to understand or to modify it without being detected by integrity checks. 

So much for theory. The practice supports several variations, which have been made throughout history. One of the most spectacular I have found during the reading of Security Engineering , of Ross Anderson. To illustrate the dissertation on man-in-the-middle attacks, the author tells the disaster suffered by the South African troops because of an attack variant that the time has come calling MIG-in-the-middle. 

Article's author note: As Zooko commented, Ross Anderson admitted in the second edition of his book that he recently discovered this history is unfounded. However it could have happened (change the actors) in other wars in other places, so I think it's still interesting.

In the late 80, South Africa was involved in a war in northern Namibia and southern Angola. His goal was to keep the white government of Namibia and to the puppet government of UNITA in Angola. As the South African army was composed exclusively of white recruits and these did not represent a very large population segment, it was essential to maintain a casualtie rate as low as possible.So the South African army preferred to concentrate on maintaining order in Namibia and support with its air force to the UNITA troops in Angola. To prevent the Angolan rebel troops and their Cuban allies to counter strike against Namibia, South Africa deployed a powerful barrier of anti-aircraft batteries. To prevent the so-called "friendly fire," South African aircraft were equipped with IFF (Identify-friend-or-foe), common today, which responded according to a predetermined algorithm to the signals from anti-aircraft batteries so that these identify themselves as friends to the South African aircraft.

The defensive system seemed unbeatable, and yet one day a squad of Cuban MiGs crossed the line of defense and bombed one of the main camps of Namibia. South African anti-aircraft batteries didn't shoot a single missile. Casualties were terrible and led the South African government to decide to withdraw troops from Namibia, which eventually led to the fall the white government of that country. 

Time passed until the discovery (and confession) of the causes of that monumental failure in the South African defense. It appears that the Angolan intelligence learned that the South Africans, perhaps due to overconfidence, left switched on their IFF devices while conducting missions over Angolan territory. What they did was to situate a MIG waiting near the border of Namibia to a South African squadron (SAAF in the figure) headed to bomb Angola.When South Africans were present on Angola to carry out their respective missions, the MIGs entered the territory of Namibia. Of course, the South African air defenses detected the MIGs and launched the IFF signals to determine whether they were friend or foe aircrafts. The MIGs lacked the necessary keys to correctly answer the IFF signals, so their plan was to forward the signals to their radio stations in Angola and make them send that signals to the South African squadron which was flying over that country at that time. The plan worked because South African airplanes had turned on their IFF and when Angolan radio stations broadcast the signals sent by the MIGs, the South Africans began to reply to them automatically without knowing that these responses were being forwarded to the MIG squadron so that they can serve it to  the South African batteries in Namibia. Thus, South African air defenses in Namibia thought that those MIGs were friendly aircrafts, letting them pass. 

That attack forced them to rethink deeply technology, protocols and operating procedures used for the IFF. To begin with, at the operational level South African prohibited pilots to let their IFFs on when they flew over enemy territory. At the protocol level the solution was much harder to find and certainly was not definitive, as they began include the aircraft identification in the response returned by the IFF so that they could correlate the data offered by control towers to find differences. Another method tried was to measure the delays of the responses so that if any came "later" than usual they could suspect that the IFF was generated by a plane that was farther away than the plane detected. Keep in mind that at this particular attack the encryption of communications is not useful as MiGs did not need to break the confidenciality or integrity of IFF signals, since they needed only to "reflect them" for their radio stations . All these answers are far from perfect and much less viable in battle areas crowded with air traffic of the two sides, which is why IFF techniques is a field in constant development to avoid disasters like that caused by the attack MIG- in-the-middle.