# US 2019/0172510 A1 · Method of training drive strength, odt of memory device, computing system performing the same and system-on-chip performing the same

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0172510 A1
- Document ID: US20190172510A1
- Status: Published application
- Kind: A1
- Application number: 16/116,369
- Filed: 2018-08-29
- Published: 2019-06-06
- Earliest claimed: 2017-12-05
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Yong-Seob KIM; Jung-Il LEE
- Web page: https://patentgrep.com/patent/US20190172510A1

## CPC

- G11C 7/222: {Clock generating, synchronizing or distributing circuits within memory device}
- G11C 5/147: {Voltage reference generators, voltage or current regulators; Internally lowered supply levels; Compensation for voltage drops (G11C5/141 takes precedence)}
- G11C 7/1066: {Output synchronization}
- G11C 7/1093: {Input synchronization}
- G11C 7/20: Memory cell initialisation circuits, e.g. when powering up or down, memory clear, latent image memory
- G11C 29/023: {in clock generator or timing circuitry}
- G11C 29/028: {with adaption or trimming of parameters}
- G11C 29/08: Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
- G11C 7/1045: {Read-write mode select circuits}
- G11C 7/1057: {Data output buffers, e.g. comprising level conversion circuits, circuits for adapting load}
- G11C 7/1084: {Data input buffers, e.g. comprising level conversion circuits, circuits for adapting load}
- G11C 2207/2254: Calibration
- H03K 19/017545: {Coupling arrangements; Impedance matching circuits}
- G11C: STATIC STORES (semiconductor memory devices H10B)
- H03K: PULSE TECHNIQUE (measuring pulse characteristics G01R; modulating sinusoidal oscillations with pulses H03C; transmission of digital information H04L; discriminator circuits detecting phase difference between two signals by counting or integrating cycles of oscillation H03D3/04; automatic control, starting, synchronisation or stabilisation of generators of electronic oscillations or pulses where the type of generator is irrelevant or unspecified H03L; coding, decoding or code conversion, in general H03M)

## Abstract

In a method of training for a memory device, an initialization operation is performed on the memory device when the memory device is powered on. A training operation is performed on a plurality of operating frequencies of the memory device such that at least one of a plurality of operating parameters of the memory device is obtained as a configurable operating parameter for each of the plurality of operating frequencies. The configurable operating parameter for each of the plurality of operating frequencies is stored as training data. An optimized operating parameter for the memory device is used based on the training data, a current operation mode of the memory device, and a current operating frequency of the memory device.

## Claims

### Claim 1

A method of training for a memory device, the method comprising:
performing an initialization operation on the memory device based on the memory device being powered on;
performing a training operation on a plurality of operating frequencies of the memory device to obtain, as a configurable operating parameter for each of the plurality of operating frequencies, at least one of a plurality of operating parameters of the memory device;
storing, as training data, the obtained configurable operating parameter for each of the plurality of operating frequencies; and
using an optimized operating parameter for the memory device based on the training data, a current operation mode of the memory device, and a current operating frequency of the memory device.

### Claim 2 (depends on 1)

The method of claim 1, wherein:
the plurality of operating parameters comprises a plurality of drive strengths of the memory device; and
the performing the training operation comprises:
measuring a plurality of data valid windows of the memory device for the plurality of operating frequencies and for the plurality of drive strengths; and
selecting at least one of the plurality of measured data valid windows based on a predetermined screen guide.

### Claim 3 (depends on 2)

The method of claim 2, wherein measuring the plurality of data valid windows comprises:
measuring first ranges of the plurality of data valid windows based on a data signal and a data strobe signal, each of the first ranges being along a first direction; and
measuring second ranges of the plurality of data valid windows based on a reference voltage, each of the second ranges being along a second direction crossing the first direction.

### Claim 4 (depends on 3)

The method of claim 3, wherein the measuring the first ranges of the plurality of data valid windows comprises fixing one of the data signal and the data strobe signal and shifting an other of the data signal and the data strobe signal.

### Claim 5 (depends on 3)

The method of claim 3, wherein the measuring the second ranges of the plurality of data valid windows comprises shifting a level of the reference voltage.

### Claim 6 (depends on 2)

The method of claim 2, wherein the storing the obtained configurable operating parameter for each of the plurality of operating frequencies comprises:
storing, as the training data, a relationship between the plurality of operating frequencies and configurable drive strengths for the plurality of operating frequencies, based on the at least one selected data valid window.

### Claim 7 (depends on 6)

The method of claim 6, wherein:
the memory device includes a plurality of operation modes;
the plurality of operation modes comprises a low power mode; and
based on the current operation mode being the low power mode, a minimum drive strength among the configurable drive strengths is determined as the optimized operating parameter.

### Claim 8 (depends on 6)

The method of claim 6, wherein:
the memory device includes a plurality of operation modes;
the plurality of operation modes comprises a high performance mode; and
based on the current operation mode being the high performance mode, a maximum drive strength among the configurable drive strengths is determined as the optimized operating parameter.

### Claim 9 (depends on 1)

The method of claim 1, wherein:
the plurality of operating parameters comprises a plurality of on-die termination (ODT) resistance values of the memory device; and
the performing the training operation comprises:
measuring a plurality of data valid windows of the memory device for the plurality of operating frequencies and for the plurality of ODT resistance values, and
selecting at least one of the plurality of measured data valid windows based on a predetermined screen guide.

### Claim 10 (depends on 1)

The method of claim 1, wherein the plurality of operating parameters comprises a plurality of drive strengths of the memory device and a plurality of ODT resistance values of the memory device.

### Claim 11 (depends on 1)

The method of claim 1, further comprising:
determining whether the memory device is cold-booted in which the initialization operation is performed for a first time or the memory device is rebooted in which the initialization operation is not performed for the first time; and
based on the memory device being determined to be rebooted, loading the training data that was previously obtained and stored,
wherein, based on the memory device being determined to be rebooted, the optimized operating parameter is used based on the training data that was previously obtained and stored.

### Claim 12 (depends on 11)

The method of claim 11, wherein, based on the memory device being determined to be rebooted, the performing the training operation and the storing of the obtained configurable operating parameter for each of the plurality of operating frequencies are not performed.

### Claim 13 (depends on 11)

The method of claim 11, further comprising:
based on the memory device being determined to be rebooted, determining whether the training data that was previously obtained and stored is normal,
wherein, based on the memory device being determined to be rebooted and the training data being determined to be normal, the optimized operating parameter is used based on the training data that was previously obtained and stored.

### Claim 14 (depends on 11)

The method of claim 11, further comprising:
based on the memory device being rebooted, determining whether a retention time of the training data is shorter than a reference time,
wherein, based on the memory device being determined to be rebooted and the retention time of the training data being determined to be shorter than the reference time, the optimized operating parameter is used based on the training data that was previously obtained and stored.

### Claim 15

(canceled)

### Claim 16

A computing system comprising:
a memory device comprising a plurality of memory cells and configured to operate based on one of a plurality of operating frequencies; and
a host configured to:
perform an initialization operation on the memory device based on the memory device being powered on,
perform a training operation on the plurality of operating frequencies to obtain, as a configurable operating parameter for each of the plurality of operating frequencies, at least one of a plurality of operating parameters of the memory device,
store, as training data, the obtained configurable operating parameter for each of the plurality of operating frequencies, and
use an optimized operating parameter for the memory device based on the training data, a current operation mode of the memory device, and a current operating frequency of the memory device.

### Claim 17 (depends on 16)

The computing system of claim 16, wherein:
the plurality of operating parameters comprises a plurality of drive strengths of the memory device; and
the host is further configured to perform the training operation by measuring a plurality of data valid windows of the memory device for the plurality of operating frequencies and for the plurality of drive strengths, and by selecting at least one of the plurality of measured data valid windows based on a predetermined screen guide.

### Claim 18 (depends on 17)

The computing system of claim 17, wherein:
the host is further configured to provide a clock signal to the memory device;
first ranges of the plurality of data valid windows are measured, based on the clock signal, by fixing one of a data signal and a data strobe signal of the memory device and by shifting an other of the data signal and the data strobe signal; and
each of the first ranges is along a first direction.

### Claim 19 (depends on 18)

The computing system of claim 18, further comprising:
a power supply device configured to provide a power supply voltage to the memory device,
wherein second ranges of the plurality of data valid windows are measured by shifting a level of a reference voltage that is generated based on the power supply voltage, and
wherein each of the second ranges is along a second direction crossing the first direction.

### Claim 20 (depends on 16)

The computing system of claim 16, wherein:
the plurality of operating parameters comprises a plurality of on-die termination (ODT) resistance values of the memory device; and
the host is further configured to perform the training operation by measuring a plurality of data valid windows of the memory device for the plurality of operating frequencies and for the plurality of ODT resistance values, and by selecting at least one of the plurality of measured data valid windows based on a predetermined screen guide.

### Claim at position 21

21-25. (canceled)

### Claim 26

A system-on-chip (SOC) comprising:
one or more integrated circuits (ICs) configured to:
perform an initialization operation on an external memory device based on the external memory device being powered on,
perform a training operation on a plurality of operating frequencies of the external memory device to obtain, as a configurable operating parameter for each of the plurality of operating frequencies, at least one of a plurality of operating parameters of the external memory device,
store, as training data, the configurable operating parameter for each of the plurality of operating frequencies, and
use an optimized operating parameter for the external memory device based on the training data, a current operation mode of the external memory device, and a current operating frequency of the external memory device.

### Claim at position 23

27-30. (canceled)
