Self-encrypting disk

According to patent #: US20110264925, self-encrypting disks consist of data, an encryption key (located among the data), a processor, instructions to be executed, and a communication interface ^[8] . In self-encrypted disks, an encryption key is generated inside the disk and never gets into the memory and the central processing unit (CPU) of the computer ^[9] . Self-encrypted disks using the AES encryption algorithm use 128 or 256 bit keys to encrypt data ^[9] .

TCG claim that the encryption process itself is constant and decryption, as well as encryption, is imperceptible to the user and the operating system ^[10]. However, to complete the work with the disk, the user needs to go through pre-boot authentication. The authentication process is inseparable from the disk. Different companies use different authentication systems, such as ATA Security or TSG OPAL ^[11] . According to TCG, self-encrypted drives using their OPAL specifications support multi-user access and multi-user administration, each with its own password and own credentials ^[12] .

According to patent #: US20120254602, preloading consists of a pre-boot operating system (OS) , unlocking software (software) , a self-encrypted disk management software, a graphical interface, an access control system, and other means (for more information, click here ).

The operation of preloading a self-encrypting disk is determined by the following order:

1. When first loaded, this software prompts the user to enter the nominal credentials and generates a disk session key ( eng. Disk session key - DSK ) that encrypts this user data.
2. Further, this software hashes the nominal user credentials and encrypts them with the DSK, which it then saves.
3. When you turn off the power, encryption is activated.
4. When the user starts the computer, the management software of the self-encrypting disk requests the nominal credentials.
5. After receiving the data, this software hashes them and decrypts the DSK key, which then decrypts the previously recorded nominal credentials.
6. Software Verifies the compliance of previously recorded and received data.
   1. In the event of inconsistency, the user is given several more (a predetermined number) of chances to enter the correct nominal credentials.
   2. If the entered data is correct, a signal is sent to decrypt the data on the self-encrypted disk. ^[13]

According to patent #: US20120254602, the management software of the self-encrypting disk deletes the encryption key when it receives a wipe command, thus the data cannot be decrypted and unavailable ^[13] . On the other hand, according to patent #: US20110264925, when a user wants to clean a disk, he is offered several cleaning options (for example: deleting the encryption key (quicker but less secure) or replacing data with others) ^[8] . TCG claims that their specification clears the existing encryption key when the disk is cleared and generates a new key, so information that was previously recorded becomes unavailable ^[14] . The above method of cipher-cleaning ( eng. Crypto-erase ) is an effective way for self-encrypting disks ^[15] .

Many self-encrypted drives use the conventional AES encryption algorithm. It uses 128 or 256 bit keys. Since 2003, the NSA considers this algorithm sufficiently stable to protect state secrets ^[16] . In the case of a self-encrypted disk, the key is stored on the disk itself in an encrypted form, so the cracker cannot take possession of it.

Since encryption occurs at the hardware level, and the encryption key never enters the memory and processor of the computer, it is impossible to conduct regular attacks through the computer's operating system, the memory of the user's computer, such as cold boot and Evil Maid attack. Moreover, after rebooting or entering sleep mode, the disk again asks for an authorization password to continue. ^[one]

There are several vulnerabilities to self-encrypting disks:

1. In spite of the security of the self-encrypted disk by the AES algorithm, the weak point is user authentication with a password.
2. Researchers at the University of Erlangen - Nuremberg found ways to hack self-encrypted disks. One of them is called Hot Plug attack . In this method, the attacked disk after authorization must be reconnected to the hacker's computer without turning off the disk power ^[11] . Other attacks are adapted from existing methods for certain situations ^[17] .

• Disk encryption
• Hardware encryption
• Hardware full disk encryption

• Tilo Müller, Tobias Latzo, and Felix C. Freiling. Self-Encrypting Disks pose Self-Decrypting Risks (English) . Hardware-based Full Disk Encryption (In) Security . Friedrich-Alexander University Erlangen-Nürnberg (2012). The appeal date is December 21, 2015.
• Trusted Computing Group. 10 Reasons to Buy Self-Encrypting Drives (English) . Trusted Computing Group (2010). The appeal date is December 21, 2015.
• Leonard E. Russo, Valiuddin Ali, Jennifer Rios, Lan Wang. Securing data on a self-encrypting storage device (English) . United States Patent and Trademark Office (October 27, 2011). The appeal date is December 21, 2015.
• Apurva M. Bhansali, Mehul R. Patel, Kamal M. Dhanani, Rajnish S. Chauhan, David Cheung. Methods, Systems, and Apparatuses for Managing a Hard Drive Security System (Eng.) . United States Patent and Trademark Office (2012-10-4). The appeal date is December 21, 2015.
• Thomas Coughlin. Solid Security: The Rise of Self-Encrypting Solid State Drives (Eng.) . SNIA Solid State Storage Initiative (2011). The appeal date is December 21, 2015.