# US 2021/0344601 A1 · Congestion Control Method, Apparatus, and System, and Computer Storage Medium

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

- Number: US 2021/0344601 A1
- Document ID: US20210344601A1
- Status: Published application
- Kind: A1
- Application number: 17/244,256
- Filed: 2021-04-29
- Published: 2021-11-04
- Earliest claimed: 2020-04-29
- Assignees: Huawei Technologies Co., Ltd.
- Applicants: Huawei Technologies Co., Ltd.
- Inventors: Haonan Ye; Liang Zhang; Jian Cheng; Jun Wu
- Web page: https://patentgrep.com/patent/US20210344601A1

## CPC

- H04L 47/33: using forward notification
- H04L 41/14: Network analysis or design
- H04L 47/11: Identifying congestion
- H04L 47/12: Avoiding congestion; Recovering from congestion
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A network device inputs first network status information of the network device in a first time period to an ECN inference model, to obtain an inference result that is output by the ECN inference model based on the first network status information. Then, the network device sends an ECN parameter sample to an analysis device that manages the network device, where the ECN parameter sample includes the first network status information and a target ECN configuration parameter corresponding to the first network status information, and the target ECN configuration parameter is obtained based on the inference result. The network device receives an updated ECN inference model sent by the analysis device.

## Claims

### Claim 1

A method implemented by a network device and comprising:
inputting first network status information of the network device in a first time period into an explicit congestion notification (ECN) inference model;
obtaining an inference result from the ECN inference model based on the first network status information, wherein the inference result comprises an original ECN configuration parameter and a confidence of the original ECN configuration parameter;
sending, to an analysis device that manages the network device, an ECN parameter sample comprising the first network status information and a target ECN configuration parameter corresponding, wherein the target ECN configuration parameter corresponds to the first network status information and is based on the inference result; and
receiving, from the analysis device, an updated ECN inference model based on training with the ECN parameter sample.

### Claim 2 (depends on 1)

The method of claim 1, further comprising performing congestion control in a second time period and using the target ECN configuration parameter, wherein the second time period is later than the first time period in a time sequence.

### Claim 3 (depends on 1)

The method of claim 1, wherein after obtaining the inference result and when the confidence is less than a confidence threshold, the method further comprises:
adjusting, based on a change of transmission performance of the network device, an ECN configuration parameter to obtain an adjusted ECN configuration parameter; and
using the adjusted ECN configuration parameter as the target ECN configuration parameter.

### Claim 4 (depends on 3)

The method of claim 3, wherein adjusting the ECN configuration parameter comprises:
when transmission performance of the network device improves from a third time period to the first time period, increasing an ECN threshold in the ECN configuration parameter, wherein the third time period is earlier than the first time period in a time sequence; and
when the transmission performance deteriorates from the third time period to the first time period, lowering the ECN threshold or increasing the ECN marking probability.

### Claim 5 (depends on 4)

The method of claim 4, wherein when a bandwidth utilization of the network device increases, a queue depth of the network device decreases, or an ECN packet ratio of the network device decreases from the third time period to the first time period, the method further comprises determining that transmission performance of the network device has improved from the third time period to the first time period.

### Claim 6 (depends on 1)

The method of claim 1, wherein after obtaining the inference result and when the confidence is greater than or equal to a confidence threshold, the method further comprises using the original ECN configuration parameter as the target ECN configuration parameter.

### Claim 7 (depends on 1)

The method of claim 1, wherein after obtaining the inference result and when the confidence is less than a confidence threshold, the method further comprises sending target indication information to the analysis device, wherein the target indication information comprises an identifier of the network device and indicates that the ECN inference model does not adapt to the network device.

### Claim 8 (depends on 7)

The method of claim 7, wherein the target indication information comprises the confidence.

### Claim 9 (depends on 1)

The method of claim 1, wherein after receiving the updated ECN inference model, the method further comprises updating the ECN inference model using the updated ECN inference model.

### Claim 10 (depends on 1)

The method of claim 1, wherein the first network status information comprises queue information, throughput information, or congestion information of the network device in the first time period.

### Claim 11

A method implemented by an analysis device and comprising:
managing a network device;
receiving, from the network device, an explicit congestion notification (ECN) parameter samples comprising network status information of the network device and a target ECN configuration parameter corresponding to the network status information;
training with the ECN parameter sample to obtain a first ECN inference model; and
sending the first ECN inference model to the network device.

### Claim 12 (depends on 11)

The method of claim 11, further comprising further training with the ECN parameter sample when an ECN inference model update condition is met, wherein the ECN inference model update condition comprises a first quantity of network devices that send target indication information to the analysis device reaching a device quantity threshold, a second quantity of times the analysis device receives the target indication information within a first duration reaching a first quantity threshold, or a third quantity of times the analysis device receives the target indication information from any network device managed by the analysis device within a second duration reaching a second quantity threshold, wherein the target indication information comprises an identifier of the network device and indicates that the ECN inference model does not adapt to the network device.

### Claim 13 (depends on 12)

The of claim 12, further comprising further sending the first ECN inference model when receiving the target indication information from the network device.

### Claim 14 (depends on 11)

The method of claim 11, wherein after training with the ECN parameter sample, the method further comprises sending, to a cloud device, an ECN message comprising the first ECN inference model and comprising an identifier of a service type corresponding to the first ECN inference model.

### Claim 15 (depends on 14)

The method of claim 14, wherein the ECN message further comprises networking information of a network managed by the network device.

### Claim 16 (depends on 11)

The method of claim 11, further comprising:
sending, to cloud device, an ECN inference model obtaining request comprising a type of a target service carried in a network managed by the analysis device;
receiving, from the cloud device, a second ECN inference model corresponding to the target service type; and
sending, to the network device, the second ECN inference model.

### Claim 17

A network device comprising:
a memory configured to store instructions; and
a processor coupled to the memory and configured to execute the instructions to:
input first network status information of the network device in a first time period into an explicit congestion notification (ECN) inference model;
obtain an inference result from the ECN inference model based on the first network status information, wherein the inference result comprises an original ECN configuration parameter and a confidence of the original ECN configuration parameter;
send, to an analysis device that manages the network device, an ECN parameter sample comprising the first network status information and a target ECN configuration parameter, wherein the target ECN configuration parameter corresponds to the first network status information and is based on the inference result; and
receive, from the analysis device, an updated ECN inference model based on training with the ECN parameter sample.

### Claim 18 (depends on 17)

The network device of claim 17, wherein the processor is further configured to perform congestion control in a second time period and using the target ECN configuration parameter, wherein the second time period is later than the first time period in a time sequence.

### Claim 19 (depends on 17)

The network device of claim 17, wherein after obtaining the inference result and when the confidence is less than a confidence threshold, the processor is further configured to:
adjust, based on a change of transmission performance of the network device, an ECN configuration parameter to obtain an adjusted ECN configuration parameter; and
use the adjusted ECN configuration parameter as the target ECN configuration parameter.

### Claim 20 (depends on 9)

The network device of claim 9, wherein the processor is further configured to adjust the ECN configuration parameter by:
when a transmission performance of the network device improves from a third time period to the first time period, increasing an ECN threshold in the ECN configuration parameter or lowering an ECN marking probability in the ECN configuration parameter, wherein the third time period is earlier than the first time period in a time sequence; and
when the transmission performance deteriorates from the third time period to the first time period, lowering the ECN threshold or increasing the ECN marking probability.

### Claim 21 (depends on 20)

The network device of claim 20, wherein when a bandwidth utilization of the network device increases, a queue depth of the network device decreases or an ECN packet ratio of the network device decreases from the third time period to the first time period, the processor is further configured to determine that the transmission performance of the network device has improved from the third time period to the first time period.

### Claim 22 (depends on 17)

The network device of claim 17, wherein after obtaining the inference result and when the confidence is greater than or equal to a confidence threshold, the processor is further configured to use the original ECN configuration parameter as the target ECN configuration parameter.

### Claim 23 (depends on 17)

The network device of claim 17, wherein after obtaining the inference result and when the confidence is less than a confidence threshold, the processor is further configured to send target indication information to the analysis device, wherein the target indication information comprises an identifier of the network device and indicates that the ECN inference model does not adapt to the network device.

### Claim 24 (depends on 23)

The network device of claim 23, wherein the target indication information comprises the confidence.

### Claim 25 (depends on 17)

The network device of claim 17, wherein after receiving the updated ECN inference model, the processor is further configured to update the ECN inference model using the updated ECN inference model.

### Claim 26

An analysis device comprising:
a memory configured to store instructions; and
a processor coupled to the memory and configured to execute the instructions to:
manage a network device;
receive, from the network device, an explicit congestion notification (ECN) parameter sample comprising network status information of the network device and a target ECN configuration parameter corresponding to the network status information;
train with the ECN parameter sample to obtain a first ECN inference model; and
send the first ECN inference model to the network device.

### Claim 27 (depends on 26)

The analysis device of claim 26, wherein the processor is further configured to further train with the ECN parameter sample when an ECN inference model update condition is met, wherein the ECN inference model update condition comprises a first quantity of network devices that send target indication information to the analysis device reaching a device quantity threshold, a second quantity of times the analysis device receives the target indication information within a first duration reaching a first quantity threshold, or a third quantity of times the analysis device receives the target indication information from any network device managed by the analysis device within a second duration reaching a second duration reaching a second quantity threshold, wherein the target indication information comprises an identifier of the network device and indicates that the ECN inference model does not adapt to the network device.

### Claim 28 (depends on 27)

The analysis device of claim 27, wherein the processor is further configured to further send the first ECN inference model when receiving the target indication information from the network device.

### Claim 29 (depends on 26)

The analysis device of claim 26, wherein after training with the ECN parameter sample, the processor is further configured to send, to a cloud device, an ECN message comprising the first ECN inference model and comprising an identifier of a service type corresponding to the first ECN inference model.

### Claim 30 (depends on 26)

The analysis device of claim 26, wherein the processor is further configure to:
send, to a cloud device, an ECN inference model comprising a type of a target service carried in a network managed by the analysis device; and
receive, from the cloud device, a second ECN inference model corresponding to the target service type; and
send, to the network device, the second ECN inference model.