Breakthrough calculations on the capacity of a steganographic channel


Steganography is the art of hiding a message in such a way that only the sender and receiver realise it is there. (By contrast, cryptography disguises the content of a message but makes no attempt to hide it.)

The central problem for steganographers is how much data can be hidden without being detected. But the complexity of the problem means it has been largely ignored in favor of more easily solved conundrums.

Jeremiah Harmsen from Google Inc in Mountain View and William Pearlman at Rensselaer Polytechnic Institute in Troy NY, say: “while false alarms and missed signals have rightfully dominated the steganalysis literature, very little is known about the amount of information that can be sent past these algorithms.”

So the pair have taken an important step to change that. Their approach is to think along the same lines as Claude Shannon in his famous determination of the capacity of a noisy channel. In Shannon’s theory, a transmission is considered successful if the decoder properly determines which message the encoder has sent. In the stego-channel, a transmission is successful if the decoder properly determines the sent message without anybody else detecting its presence.

Previous attempts have all placed limits on the steganographers channel for example, by stipulating that the hidden data, or stego-channel, has the same distribution as the cover channel. But Harmsen and Pearlman have take a more general approach which takes some important steps towards working out the channel capacity over a much wider range of conditions.

The results are interesting and in some cases counter-intuitive (for example, adding noise to channel can increase its steganographic capacity and in some cases, mounting two attacks on a channel instead of one can do the same).

It’s fair to say that Harmsen and Pearlman are pioneering of the study of steganographic capacity and that with this breakthrough, the field looks rich with low hanging fruit. Expect more!

Ref: Capacity of Steganographic Channels

7 Responses to “Breakthrough calculations on the capacity of a steganographic channel”

  1. l a says:

    I resent the term pioneering, especially since steganography has been around since the dark ages…maybe they are pioneering improving on
    electronic steganography, but as a whole….there have been many before them…check the freemasons for a good example.

  2. B a says:

    It doesn’t say pioneering in steganography it says “…pioneering of the study of steganographic capacity.” The whole article is about how the study of capacity has never gotten much attention. I think your resentment is misdirected, unless you think the capacity conundrum is already well addressed.

  3. Reow says:

    I resent your resentment. You are a n00b with poor reading skills.

  4. Young Blue says:

    I resent poor grammer:

    “…Harmsen and Pearlman are pioneering of the study…”


  5. Alex says:

    Very interesting result. I assume it’s an entropic measurement as well?

  6. jack sprat says:

    and i resent the resenter of the resentment.

  7. anomaly says:

    I resent that the resenter resents the resenter of the resentment. Resent that.