UPI Intelligence Correspondent

The National Security Agency has mounted fresh efforts to crack the top-secret code of Iran's Ministry of Intelligence and Security, following the NSA's discovery of an alleged Iranian agent working in close cooperation with the Bush administration in Iraq, according to serving intelligence officials.

The MOIS Baghdad station chief was alerted that the United States was allegedly reading its mail by Pentagon protégé, Ahmed Chalabi, an Iraqi exile, according to United Press International's news accounts.

Chalabi, the subject of a full field FBI investigation, has strongly denied any wrongdoing.

The NSA became aware of Chalabi's alleged warning by means of an intercept sent by the MOIS station chief to Tehran, U.S. officials said.

The response of the NSA in Ft. Meade, Md. -- which has 18 acres of enormously powerful Cray-1 supercomputers capable of transferring 320 million words per second -- was to relaunch efforts to crack the Iranian code, U.S. intelligence officials said.

But this time, they are encountering difficulty. According to one former NSA official, Iran is "staying off the air," and is avoiding U.S. technical intercepts by resorting to a much older system that involves "one-time pads."

Milt Bearden, former chief of CIA operations in Afghanistan, when informed of the Iranian tactic was not surprised. "The Iranians were messing around with codes and secret writing when Americans were still smearing themselves with blue ochre," he said.

Authors John Earl Haynes and Harvey Klehr in the book "Venona: Decoding Soviet Espionage in America," interviewed NSA operatives who compromised the Soviet encryption system and they explained how a one-time pad works.

According to them, a concise message was written in a Soviet code clerk's office. The code clerk's task was to convert the message into four-digit numerical codes.

Each page of such a pad is used in conjunction with a codebook containing 60 five-digit numerical groups, called the "additive key," according to the authors.

The first numerical group appears in the upper left hand corner of the book, they said.

Haynes and Klehr used as an actual example, a report from a traitorous U.S. Air Force officer working for the KGB. Knowing that a report from the officer had arrived, his KGB handler wrote, "Pilot delivered report about rockets."

The code clerk then converted the message into numbers so that it read: "Pilot delivered report about rockets" -- "7934 2157 1139 3872 2166."

The code clerk then began to use the one-time pad. On each page of the pad are 60 five-digit numerical groups. The clerk then took the first numerical group in the upper left hand corner, 26473 and wrote it down. According to the authors, this was done to alert the receiving code clerk as to which page the enciphering clerk was using to encode his message.

The receiving code clerk would be using the same one-time pad.

The initial code clerk then placed the second five-digit group on the one-time pad beneath the numbers from the codebook. The numbers were added together. If the sum was less than 9, nothing was carried.

As the authors explained, 8 plus 6 equals 4 because the one is not carried.

So by using the codebook, the original message consisting of four-digit number group is changed to the following:

From the codebook, you would have: "79342 15711 39387 22166."

From the one-time pad, you would have: "26473 56328 29731 35682 23798."

Thus, the final enciphered message would read: "26473 25660 34442 64969 45854."

Substituting numbers for letters then converts the five digit groups to five-letter groups. Thus, Latin letter o=O, 1=I, 2=U, 3=Z, 4=T, 5=R, 6=E, 7=W, 8=A, 9=P.

Thus your final message would read: "UETWZ UREEO ZITTU ETPEP TRART.../TEERO 32412."

To decipher this, the receiving code clerk goes to his copy of the one-time pad, noting that the group 26374 is the right page to use for decoding. The clerk then subtracts the five-digit groups provided him by the one-time pad, according to Haynes and Klehr.

One former NSA official interviewed by United Press International said this is "almost impossible to break. It's terribly difficult."

Haynes and Klehr quote Cecil Philips, one of the analysts who helped break the Soviet system as saying, "The security of such an encipherment-decipherment system depends on the randomness (that is, unpredictability) of the 'key' on the one-time pad pages and the uniqueness of the one-time pad sets held by the sender and receiver."

According to the authors, the big disadvantage of the one-time pad system is the manpower required to produce the pads. They noted that "a small army of code makers" was employed by the Soviets in producing them.

They said that after Germany's Adolph Hitler invaded the Soviet Union in June of 1941, the stock of one-time pads began to melt away, and code workers who did not meet quotas faced severe penalties. So they began to cheat, duplicating key pages so that soon tens of thousands of duplicated key pages were inserted in one-time pads.

It would prove a fatal error.

According to Haynes and Klehr, Britain and the United States were able to gather enough of the duplicated pages to begin an attack on the Soviet system.

The NSA, which had huge resources and an enormous team of philologists, linguists, engineers and technicians, would prove absolutely key to the effort, which took years, according to the authors.

The discovery of the duplicate pages was made in 1943 and constituted a real breakthrough, the authors said.

The authors maintain one of the heroes of the project was Meredith Gardner, a soft-spoken Southerner who read German, Sanskrit, Lithuanian, Spanish and French. He soon mastered Russian and was able to recreate the Soviet code, crumbling the Soviet one-time pad system into total ruin.

It was then discovered that the Soviets had highly placed spies in almost every important U.S. government agency, and operatives even in the Roosevelt White House, the authors said.

It eventually led to Ethel and Julius Rosenberg who gave atomic bomb secrets to Russia and who were the only Americans ever to be executed for espionage, according to Haynes and Klehr.

But these authors added that had the Venona messages been made public, they would have exposed three other Americans who were atomic spies, and removed the Rosenbergs from the spotlight of American public outrage. 

http://www.washingtontimes.com/upi-breaking/20040827-042326-6404r.htm