Patent classifications
G11C11/565
Drive strength calibration for multi-level signaling
Methods, systems, and devices for drive strength calibration for multi-level signaling are described. A driver may be configured to have an initial drive strength and to drive an output pin of a transmitting device toward an intermediate voltage level of a multi-level modulation scheme, where the output pin is coupled with a receiving device via a channel. The receiving device may generate, and the transmitting device may receive, a feedback signal indicating a relationship between the resulting voltage of the channel and an value for the intermediate voltage level. The transmitting device may determine and configure the driver to use an adjusted drive strength for the intermediate voltage level based on the feedback signal. The driver may be calibrated (e.g., independently) for each intermediate voltage level of the multi-level modulation scheme. Further, the driver may be calibrated for the associated channel.
Semiconductor memory device, memory system, and write method
According to one embodiment, a memory system includes a semiconductor memory device including a memory cell capable of holding at least 4-bit data and a controller configured to control a first write operation and a second write operation based on the 4-bit data. The controller includes a conversion circuit configured to convert 4-bit data into 2-bit data. The semiconductor memory device includes a recovery controller configured to recover the 4-bit data based on the converted 2-bit data and data written in the memory cell by the first write operation. The first write operation is executed based on the 4-bit data received from the controller, and the second write operation is executed based on the 4-bit data recovered by the recovery controller.
DATA PROCESSING SYSTEM AND OPERATION METHOD OF DATA PROCESSING SYSTEM
To provide a data processing system that includes a nonvolatile memory device capable of storing multilevel data and enables increasing storage capacity of a main memory device when the data processing system is activated. The data processing system includes an arithmetic processing device, a main memory device, and a nonvolatile memory device. The main memory device includes a volatile memory device, and the nonvolatile memory device is configured to store multilevel data in one memory cell. When the data processing system is deactivated, the nonvolatile memory device stores binary data, whereby the stored data can be held for a long time. Upon activation, the nonvolatile memory device stores multilevel data, whereby increasing storage capacity. When the storage capacity is increased, a free space is generated in the nonvolatile memory device, which can be used as a part of the main memory device of the data processing system.
System and method for reading and writing memory management data using a non-volatile cell based register
Methods, systems, and devices for reading and writing memory management data using a non-volatile cell based register are described. A memory device may include a set of latch units addressable via a set of row lines and a set of column lines. Each latch unit may include a sense amplifier coupled with a first line and a first non-volatile capacitor coupled with the first line and a second line, where the first capacitor is configured to store a charge representing one or more bits. Additionally, each latch unit may include a second capacitor coupled with the first line and a third line, where the second capacitor is configured to amplify a voltage at the first line based on the charge stored in the first capacitor.
SEDRAM-BASED STACKED CACHE SYSTEM AND DEVICE AND CONTROLLING METHOD THEREFOR
The present disclosure relates to a SEDRAM-based stacked Cache system, a controlling method and a Cache device. The stacked Cache system is integrated in multiple layers of wafers bonded together and includes a Cache, a Cache controller and a SEDRAM controller; the multiple layers of wafers includes a SEDRAM wafer structure and a processor wafer structure; a SEDRAM unit is integrated in each layer of SEDRAM wafer in the SEDRAM wafer structure and configured as a storage space of the Cache; a CPU, the Cache controller, the SEDRAM controller and a memory controller are integrated in the processor wafer structure.
Sensing techniques using a charge transfer device
Techniques are provided for sensing a signal associated with a memory cell capable of storing three or more logic states. To sense the memory cell (e.g., a signal associated with the memory cell), a charge may be transferred between a digit line and a node coupled with a plurality of sense components using a charge transfer device. Once the charge is transferred, at least some if not each of the plurality of sense components may sense the charge using one of a variety of sensing schemes. For example, the charge may be sensed by each sense component at a same time using a single fixed reference value, or at different times using different fixed reference values. Based on the charge being transferred or transferred with the node (e.g., using the charge transfer device) and each sense component sensing the charge, a logic state associated with the memory cell may be determined.
Semiconductor device performing a multiplication and accumulation operation
A semiconductor device includes a cell array, a computation circuit, and a control circuit. The cell array includes a plurality of unit cells configured to store a plurality of first signals by a write operation and to output a plurality of output signals corresponding to the first signals by a read operation. The computation circuit includes a plurality of unit computation circuits receiving the plurality of output signals and being set according to a plurality of second signals during a computation operation. The control circuit is configured to control the cell array and the computation circuit during the write operation, the read operation, and the computation operation.
Data processing system and operation method of data processing system
To provide a data processing system that includes a nonvolatile memory device capable of storing multilevel data and enables increasing storage capacity of a main memory device when the data processing system is activated. The data processing system includes an arithmetic processing device, a main memory device, and a nonvolatile memory device. The main memory device includes a volatile memory device, and the nonvolatile memory device is configured to store multilevel data in one memory cell. When the data processing system is deactivated, the nonvolatile memory device stores binary data, whereby the stored data can be held for a long time. Upon activation, the nonvolatile memory device stores multilevel data, whereby increasing storage capacity. When the storage capacity is increased, a free space is generated in the nonvolatile memory device, which can be used as a part of the main memory device of the data processing system.
COMPUTING-IN-MEMORY ACCELERATOR DESIGN WITH DYNAMIC ANALOG RAM CELL AND ASSOCIATED LOW POWER TECHNIQUES WITH SPARSITY MANAGEMENT
Systems formed by a multi-bit three-transistor (3T) memory cell (i.e., dynamic-analog RAM) are provided. The 3T memory cell includes: a read-access transistor M.sub.1 in electrical communication with a read bitline; a switch transistor M.sub.2 in electrical communication with the read-access transistor M.sub.1 a write-access transistor M.sub.3 in electrical communication with the read-access transistor M.sub.1 and a write bitline; and a memory node MEM in electrical communication between the read-access transistor M.sub.1 and the write-access transistor M.sub.3, wherein the memory node MEM is configured to store a 4-bit weight WE. An array of the 3T memory cells (i.e., dynamic-analog RAMs) may form a computing-in-memory (CIM) macro, and further form a convolutional neural network (CNN) accelerator by communicating with an application-specific integrated circuit (ASIC) which communicates with a global weight static random access memory and an activation static random access memory.
Apparatuses, systems, and methods for frequency-dependent signal modulation
Apparatuses, systems, and methods for high-pass filtering pre-emphasis circuits. A device may use a pre-emphasis driver to provide a multi-level signal based on multiple binary signals. The pre-emphasis driver includes a primary driver coupled in parallel with at least one equalizer path, each of which includes an equalizer driver and a filtering element. The filtering element may be an AC filtering element, such as a capacitor. The equalizer paths may contribute equalized signal(s) which have a high-pass filtering behavior. The pre-emphasis circuit may combine the primary signal from the primary driver and the equalized signals to generate an overall output multi-level signal. In some embodiments, the pre-emphasis driver may be a pulse amplitude modulated (PAM) driver, such as a PAM4 driver with four levels of the multi-level driver.