Samsung Electronics Co., Ltd.

Pilsang Yoon

Changkyu Seol

Hong Rak Son

Junjin Kong

Hongrak Son

Jinsoo Lim

Hyejeong So

Jun Jin Kong

Jiyoup Kim

Dongsup Jin

Eun Chu Oh

Myoungbo Kwak

Jaehun Jang

Youngdon Choi

Sucheol Lee

Junghwan Choi

Daeseok Byeon

Jisu Kim

Jinman Han

Young-Seop Shim

Seonghyeog Choi

Geunyeong Yu

Taehyun Song

Dongjin Shin

Myunghoon Choi

Jinbae Bang

Donghwan Lee

Heewon Lee

Hwaseok Oh

Kwanwoo Noh

Seonghoon Woo

Chanho Yoon

Seungkyung Ro

Donggi Lee

Jae Hun Jang

Seongnam Kwon

Seong Hyeog Choi

Daewook Kim

Jung Soo Chung

Ju-young Jung

Mankeun Seo

Jungsoo Chung

Ju-Young Jung

Junghyun Hong

Soonjae Won

Seonghyeong Choi

Encoding and decoding apparatuses and methods for implementing multi-mode coding are provided. The apparatus includes a transmitter and a receiver connected to a data bus. When data bursts are converted by the transmitter into codewords each including a plurality of symbols and/or a codeword received by the receiver is recovered as data bursts, maximum transition avoidance (MTA) codeword mappings in which no maximum transition (MT) event occurs between the plurality of symbols and minimum DC current (MDC) codeword mappings related to minimum power consumption of the plurality of symbols are used.

Encoding and decoding apparatuses and methods for implementing multi-mode coding

An electronic device includes a memory storing data from an external source, an application processing unit (APU) transmitting a secret key and public key generation command, an isolated execution environment (IEE) generating a secret key in response to the secret key generation command, generating a public key based on the secret key in response to the public key generation command, and storing the secret key, and a non-volatile memory performing write and read operations depending on a request of the APU. When the data are stored in the memory, the APU transmits a public key request to the IEE and in response the IEE transfers the public key to the APU through a mailbox protocol. The APU generates a ciphertext by performing homomorphic encryption on the data based on an encryption key in the public key, and classifies and stores the public key and the ciphertext in the non-volatile memory.

Electronic device using homomorphic encryption and encrypted data processing method thereof

Disclosed is an electronic device, which includes an ECC decoder that performs ECC decoding on a flit including a plurality of PAM-4 symbols for each of a plurality of ECC groups, a CRC decoder that performs CRC decoding on the ECC decoded flit to obtain data, and an erasure decoding unit that calculates an LLR for each of the PAM-4 symbols when the CRC decoding fails, extracts an error symbol candidate from among the plurality of PAM-4 symbols for each of the plurality of ECC groups based on the LLR, and performs the ECC decoding again after erasing the error symbol candidate.

Error correction using PAM4 modulation

Provided is a device and method for encoding and decoding to implement maximum transition avoidance coding with minimum overhead. An exemplary device performs encoding and/or decoding, by using sub-block lookup tables representing correlations between some bit values in a data burst and symbols, a combining lookup table selectively interconnecting the sub-block lookup tables based on remaining bit values of the data burst, and a codeword decoding lookup table designating the sub-block lookup tables corresponding to the symbols of each of received codewords.

Devices and methods for encoding and decoding to implement a maximum transition avoidance coding with minimum overhead

A neuromorphic package device includes a systolic array package and a controller. The systolic array package includes neuromorphic chips arranged in a systolic array along a first direction and a second direction. The controller communicates with a host controls the neuromorphic chips. Each of the neuromorphic chips sequentially transfers weights of a plurality layers of a neural network system in the first direction to store the weights. A first neuromorphic chip performs a calculation based on stored weights therein and an input data received in the second direction, and provides a result of the calculation to at least one of a second neuromorphic chip and a third neuromorphic chip which are adjacent to the first neuromorphic chip. The at least one of the second and third neuromorphic chips performs a calculation based on a provided result of the calculation and stored weights therein.

Neuromorphic package devices and neuromorphic computing systems

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.

Stacked neuromorphic devices and neuromorphic computing systems

A storage device is provided including a memory controller having a neural processing unit (NPU); a first nonvolatile memory (NVM) connected to the memory controller through a first channel; and a second NVM connected to the memory controller through a second channel. The first NVM stores first weight data for the NPU and the second stores second weight data for the NPU. The memory controller is configured to determine one of the first and second channels that is less frequently accessed upon receiving an inference request from the neural processor, and access a corresponding one of the first weight data and the second weight data using the determined one channel.

Storage device and operating method of storage device

A storage system includes a host device and a storage device. The host device provides first input data for data storage function and second input data for artificial intelligence (AI) function. The storage device stores the first input data from the host device, and performs AI calculation based on the second input data to generate calculation result data. The storage device includes a first processor, a first nonvolatile memory, a second processor and a second nonvolatile memory. The first processor controls an operation of the storage device. The first nonvolatile memory stores the first input data. The second processor performs the AI calculation, and is distinguished from the first processor. The second nonvolatile memory stores weight data associated with the AI calculation, and is distinguished from the first nonvolatile memory.

Storage device with artificial intelligence and storage system including the same

A neuromorphic device includes a neuron block, a spike transmission circuit and a spike reception circuit. The neuron block includes a plurality of neurons connected by a plurality of synapses to perform generation and operation of spikes. The spike transmission circuit generates a non-binary transmission signal based on a plurality of transmission spike signals output from the neuron block and transmits the non-binary transmission signal to a transfer channel, where the non-binary transmission signal includes information on transmission spikes included in the plurality of transmission spike signals. The spike reception circuit receives a non-binary reception signal from the transfer channel and generates a plurality of reception spike signals including reception spikes based on the non-binary reception signal to provide the plurality of reception spike signals to the neuron block, where the non-binary reception signal includes information on the reception spikes.

Neuromorphic device and neuromorphic system including the same

Channel selection information indicate positions of data bits of input data, positions of error correction code (ECC) parity bits for correcting errors in the input data, and positions of state shaping parity bits. The ECC parity bits and the state shaping parity bits are generated to cause a decrease in a quantity of memory cells, of the plurality of memory cells, in which at least one target state among a plurality of states is programmed. An alignment vector is generated based on aligning the data bits of the input data, the ECC parity bits, and the state shaping parity bits, based on the channel selection information. A codeword is generated based on simultaneously performing state shaping and ECC encoding with respect to the alignment vector. Write data are written in the nonvolatile memory device based on the codeword.

Methods of controlling operation of nonvolatile memory devices and data converters for performing the same

An method of operating a nonvolatile memory device including a plurality of memory cells comprises receiving a read command from an external device, in response to the read command, performing, based on a reference voltage, a first cell counting operation with respect to the plurality of memory cells, adjusting at least one read voltage of first through nth read voltages (where n is a natural number greater than 1) based on a first result of the first cell counting operation, and performing, based on the adjusted at least one read voltage, a read operation corresponding to the read command with respect to the plurality of memory cells.

Operation method of nonvolatile memory device and storage device

A nonvolatile memory storage system includes a plurality of memory cells and a memory controller configured to transmit a read command to a nonvolatile memory device based on a plurality of read voltages. The nonvolatile memory device performs a first read operation on a first level among the N levels based on a first read voltage among the plurality of read voltages, counts the number of on-cells that respond to the first read voltage among the plurality of memory cells, and adjusts a level of a second read voltage to be used to perform a second read operation on the first level or a second level among the N levels among the plurality of read voltages according to a comparison result of the counted number of on-cells and the number of reference cells.

Nonvolatile memory storage system

Growing community of inventors

Hwaseong-si, South Korea

Pilsang Yoon