Abstract:
A method of operating a data storage device according to an exemplary embodiment of the present inventive concepts includes generating at least one pseudo noise (PN) sequence using logical information and physical information for the data storage device, and converting first data into second data using the at least one PN sequence. Generating the at least one PN sequence includes generating a random seed based on the logical information and the physical information, and generating the at least one PN sequence using the random seed. The logical information may be a logical page address for the data storage device, and the physical information may be a physical page address for the data storage device.
Abstract:
In a method of error correction code (ECC) decoding, normal read data are read from a nonvolatile memory device based on normal read voltages, and a first ECC decoding is performed with respect to the normal read data. When the first ECC decoding results in failure, flip read data are read from the nonvolatile memory device based on flip read voltages corresponding to a flip range of a threshold voltage. Corrected read data are generated based on the flip read data by inverting error candidate bits included in the flip range among bits of the normal read data, and a second ECC decoding is performed with respect to the corrected read voltage. Error correction capability may be enhanced by retrying ECC decoding based on the corrected read data when ECC decoding based on the normal read data results in failure.
Abstract:
An encryption method used in the memory system includes; generating a private key using physical unique identification (PUID) information of a nonvolatile memory device, encrypting data using the private key, and then programming the encrypted data in the nonvolatile memory device.
Abstract:
A stacked neuromorphic device includes a logic die including a control circuit and configured to communicate with a host, and core dies stacked on the logic die and connected to the logic die via through silicon vias (TSVs) extending through the core dies. The core dies include a neuromorphic core die including a synapse array connected to row lines and column lines. The synapse array includes synapses configured to store weights and perform a calculation based on the weights and input data. The weights are included in layers of a neural network system. And the control circuit provides the weights to the neuromorphic core die through the TSVs and controls data transmission by the neuromorphic core die.
Abstract:
An encryption device includes: a parameter generating circuit configured to generate an encryption parameter including a number of initial valid bits based on an operation scenario; an encryption circuit configured to generate a cipher text by encrypting a plain text received from the outside, based on the encryption parameter; an operation circuit configured to generate a final cipher text by performing a plurality of operations on the cipher text according to the operation scenario and tag, to the final cipher text, history information of the operations performed on the final cipher text; and a decryption circuit configured to generate a decrypted plain text by decrypting the final cipher text and output a number of reliable bits of the decrypted plain text based on the history information.
Abstract:
A homomorphic encryption device includes: a recryption parameter generating circuit, a recryption circuit, and an arithmetic circuit. The recryption parameter generating circuit is configured to generate a recryption parameter including a plurality of recryption levels respectively for a plurality of ciphertexts based on an arithmetic scenario including information about an arithmetic schedule between the plurality of ciphertexts. The recryption circuit is configured to generate a plurality of recrypted ciphertexts by recrypting each of the plurality of ciphertexts to a corresponding recryption level based on the recryption parameter. The arithmetic circuit is configured to output an arithmetic result by performing operations by using the plurality of recrypted ciphertexts, according to the arithmetic scenario.
Abstract:
According to a method of controlling an operation of a nonvolatile memory device using machine learning, operating conditions of the nonvolatile memory device are determined by performing an inferring operation using a machine learning model. Training data that are generated based on feature information and error information are collected, where the error information indicate results of error correction code (ECC) decoding of the nonvolatile memory device. The machine learning model is updated by performing a learning operation based on the training data. Optimized operating conditions for individual user environments are provided by collecting training data in the storage system and performing the learning operation and the inferring operation based on the training data.