0

I have a proposed method to encrypt files (I don't know if it is actually applied or discussed). The way I suggest is to use a sector to sector table in a random manner, So if a specific sector of the file is x1 is mapped to y1 according to the sector to sector table. The sector to sector table is created once at the first time OS is created or expanded as needed. What is the benefit of this method over full-disk or file-based encryption? The sector to sector table can be considered as a very large encryption key which makes brute force almost impossible. The disadvantage needs hardware acceleration to make the sector to sector mapped seamlessly according to encrypted/mapped table, the second drawback is small file size will not spread too much. Can this idea be implemented in practice?

Improve this question

1 Answer 1

Reset to default
1

This is not encryption. The data on the disk would remain unchanged. An attacker can easily find data, e.g. by pieces of text for plain text files, by headers of binary files.

To you question Can this idea be implemented in practice? You can implement everything you want. You just should understand that it is not quite resistant against attacks.

Improve this answer
4
  • right, data itself is not encrypted but the whole file looks like encrypted by scattering randomly (without key), for text files that right it considers being small files, so not benefit from this method, but for bigger files like images and videos it will be hard to combine pieces of a file, also this method could be used with full disc or file-based encryption.
    – Computer_guy11
    Apr 26, 2020 at 13:44
  • 1
    1) Data in video are not random. Close pieces can have similar statistical values, which simplifies combining of different pieces into a sequence. 2) In many cases an attacker does not need the whole file, but only a particular piece of it.
    – mentallurg
    Apr 26, 2020 at 13:54
  • Data in a file could be statistically similar, but the process of finding those similar data take a huge time of searching
    – Computer_guy11
    Apr 26, 2020 at 14:34
  • 1
    Huge? It depends. Big data solutions like Apache Hadoop, Apache Spark, Amazon EMR etc. can process Petabytes of data in acceptable time: from a couple of hours to a couple of days. Where as to brute-force AES the computing of the whole world over millions of years would be not sufficient.
    – mentallurg
    Apr 26, 2020 at 15:51

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .