Abstract:
A method for tracing traitor receivers in a broadcast encryption system. The method includes using a false key to encode plural subsets representing receivers in the system. The subsets are derived from a tree using a Subset-Cover system, and the traitor receiver is associated with one or more compromised keys that have been obtained by a potentially cloned pirate receiver. Using a clone of the pirate receiver, the identity of the traitor receiver is determined, or the pirate receiver clones are rendered useless for decrypting data using the compromised key by generating an appropriate set of subsets.
Abstract:
An encryption key matrix has rows grouped into segments, with a set of one segment per column establishing a slot. Slots are assigned to device manufacturers, with the keys of the slots then being assigned to decryption devices made by the respective manufacturer. In generating the slots, the number “q” of segments in a column is first defined such that a predetermined maximum number of devices can be revoked devices (in that all the keys held by the device are revoked) while ensuring that a good device remains a functional device with a probability of at least (1−Q), wherein Q is a predefined device confidence. Once the number “q” of segments has been defined, the slots themselves are defined in a provably non-discriminatory fashion using an error-correcting code such as a Reed-Solomon code. With this invention, overlap between slots can be minimized to minimize the possibility that the key set of an innocent device might be inadvertently revoked when the keys in the slots of a “bad” manufacturer are revoked.
Abstract:
Sets of encryption keys useful by devices for decrypting encrypted content are defined using an error-correcting code such as a Reed-Solomon code to define vectors of length “n” over an alphabet of (0, . . . , N−1), wherein “n” is the number of columns in a key matrix and “N” is the number of rows in the matrix. Each vector represents a set of keys that can be assigned to a device. With this invention, overlap between sets of keys can be minimized to minimize the possibility that the key set of an innocent device might be inadvertently revoked when the key set of a compromised device is revoked. Also, only the generating matrix of the error-correcting code and the index of one set of keys need be stored in memory, since all previously defined key sets can be regenerated if need be from just the generating matrix and index.
Abstract:
Systems and methods for estimating data reduction ratio for a data set is provided. The method comprises selecting a plurality of m elements from a data set comprising a plurality of N elements; associating an identifier hi for each of the plurality of m elements; associating an identifier he for each of the plurality of elements in the data set; tracking number of times an element i appears in a base set that includes the plurality of m elements selected from the data set; calculating a value counti that indicates the number of times an identifier he matches an identifier hi; and estimating data reduction ratio for the plurality of N elements in the data set, based on number of m number elements selected from the data set and the value counti.
Abstract:
A tree is used to partition stateless receivers in a broadcast content encryption system into subsets. Two different methods of partitioning are disclosed. When a set of revoked receivers is identified, the revoked receivers define a relatively small cover of the non-revoked receivers by disjoint subsets. Subset keys associated with the subsets are then used to encrypt a session key that in turn is used to encrypt the broadcast content. Only non-revoked receivers can decrypt the session key and, hence, the content.
Abstract:
A tree is used to partition stateless receivers in a broadcast content encryption system into subsets. Two different methods of partitioning are disclosed. When a set of revoked receivers is identified, the revoked receivers define a relatively small cover of the non-revoked receivers by disjoint subsets. Subset keys associated with the subsets are then used to encrypt a session key that in turn is used to encrypt the broadcast content. Only non-revoked receivers can decrypt the session key and, hence, the content.
Abstract:
A tree is used to partition stateless receivers in a broadcast content encryption system into subsets. Two different methods of partitioning are disclosed. When a set of revoked receivers is identified, the revoked receivers define a relatively small cover of the non-revoked receivers by disjoint subsets. Subset keys associated with the subsets are then used to encrypt a session key that in turn is used to encrypt the broadcast content. Only non-revoked receivers can decrypt the session key and, hence, the content.
Abstract:
A method for providing a proactive security in proactive operating environment. The proactive operating environment includes a group of proactive servers communicating over a network. Each proactive server (PSI) includes a storage that includes a non erasable part that stores a public, non proactive related, key VIStart. The storage further includes an erasable part for storing private and public data. The proactive server has a discardable one-time private key SIStart that corresponds to the public key VIStart. The proactive server further has configuration data C. There is further provided a processor for providing a proactive services to applications. The proactive server has a group public proactive key VCERT common to the group of proactive servers and a share SICERT of a corresponding private proactive key SCERT. The method further includes the steps of invoking initialization procedure for generating restore related information, and invoking a restore procedure for utilizing the public, non proactive related, key VIStart and the restore related information for restoring the public proactive key VCERT.
Abstract:
A system for protecting content on recordable media for, e.g., DVD audio disks, flash memory media, or other media includes providing a media key block (MKB) on each media, with each MKB including 25,000 encryptions of a media key by 25,000 or so device keys. Each authorized player in the system has a single device key from among the system device keys with which to decrypt the media key. To avoid a coincidence attack in which a hacker can learn the MKB and associated media key and then guess at a device key without knowing its position in the MKB, the media key is XORed with a number representing each position in the MKB, and only then encrypted with the device key corresponding to that position.