# US 11,310,125 B2 · AI-enabled adaptive TCA thresholding for SLA assurance

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

- Number: US 11,310,125 B2
- Document ID: US11310125B2
- Status: Granted
- Kind: B2
- Application number: 15/989,760
- Filed: 2018-05-25
- Published: 2019-11-28
- Granted: 2022-04-19
- Earliest claimed: 2018-05-25
- Assignees: AT&T Intellectual Property I, L.P.
- Applicants: AT&T Intellectual Property I, L.P.
- Inventors: Wen-Jui Li; Tsong-Ho Wu; Jun-Min Liu
- Web page: https://patentgrep.com/patent/US11310125B2

## CPC

- H04L 41/5019: Ensuring fulfilment of SLA
- H04L 41/0681: Configuration of triggering conditions
- H04L 41/147: for predicting network behaviour
- H04L 41/16: using machine learning or artificial intelligence
- H04L 41/5016: {based on statistics of service availability, e.g. in percentage or over a given time}
- H04L 41/40: using virtualisation of network functions or resources, e.g. SDN or NFV entities
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Systems and methods for adaptive TCA threshold management are described. A technique can include receiving threshold crossing alert (TCA) data including operational TCA data, determining a TCA prediction trigger based on a current threshold, matching, in response to determining the TCA prediction trigger, the operational TCA data to a pattern from a TCA pattern bank, and calculating a predicted TCA threshold based on the pattern.

## Claims

### Claim 1

A method, comprising:
receiving, by a processing system including a processor, threshold crossing alert (TCA) data relating to a service level agreement (SLA) and including operational TCA data;
determining, by the processing system, that a threshold not being satisfied, the threshold associated with the TCA, is associated with a root cause of an SLA impact;
triggering, by the processing system, a dynamic prediction of a new TCA trigger resulting in a TCA prediction trigger based on the TCA and the determining step;
matching, by the processing system, in response to determining, by the processing system, the TCA prediction trigger, the operational TCA data to a pattern of a plurality of patterns from a TCA pattern bank, wherein the determining, by the processing system, of the TCA prediction trigger includes determining, by the processing system, from the TCA data, the service level agreement (SLA) impacts exceed a target SLA impact count associated with a current threshold and determining, by the processing system, from the TCA data, that resource utilization exceeds a resource target utilization with the current threshold;
storing, by the processing system, the plurality of patterns according to a tree data structure in the TCA pattern bank;
using, by the processing system, the tree data structure to segment a total period of monitoring into a plurality of sets to establish branches and leaves as neighbors; and
calculating, by the processing system, a predicted TCA threshold based on the pattern from the TCA pattern bank, wherein the TCA pattern bank comprises the plurality of sets.

### Claim 2 (depends on 1)

The method of claim 1, wherein the receiving of the TCA data includes retrieving, by the processing system, the TCA data from a TCA database, wherein the TCA data is historical TCA data.

### Claim 3 (depends on 1)

The method of claim 1, wherein the operational TCA data is monitored in a continuous manner.

### Claim 4 (depends on 3)

The method of claim 3, further comprising:
storing, by the processing system, the TCA data in a TCA database.

### Claim 5 (depends on 1)

The method of claim 1, wherein the TCA data includes one or more of service level agreement (SLA) TCA data and resource utilization TCA data.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
identifying, by the processing system, one or more patterns in the TCA data.

### Claim 7 (depends on 6)

The method of claim 6, wherein the TCA prediction trigger causes an adaptive thresholding process to be engaged.

### Claim 8 (depends on 1)

The method of claim 1, wherein the calculating of the predicted TCA threshold based on the pattern from the TCA pattern bank includes calculating, by the processing system, the predicted TCA threshold based on a set of neighbor thresholds associated with the pattern.

### Claim 9 (depends on 1)

The method of claim 1, wherein the matching of the operational TCA data to the pattern from the TCA pattern bank includes matching, by the processing system, based on a heuristic.

### Claim 10 (depends on 9)

The method of claim 9, wherein the heuristic includes a similarity check and a behavior range check.

### Claim 11 (depends on 1)

The method of claim 1, wherein the matching of the operational TCA data to the pattern from the TCA pattern bank includes matching, by the processing system, based on machine learning.

### Claim 12 (depends on 11)

The method of claim 11, wherein the machine learning includes k-means clustering.

### Claim 13 (depends on 1)

The method of claim 1, further comprising:
implementing, by the processing system, the predicted TCA threshold.

### Claim 14

A system, the system comprising:
a processing system including a processor;
memory coupled with the processing system, the memory comprising executable instructions that when executed by the processing system cause the processing system to effectuate operations comprising:
receiving threshold crossing alert (TCA) data relating to a service level agreement (SLA) and including operational TCA data;
determining that a threshold not being satisfied, the threshold associated with the TCA, is associated with a root cause of an SLA impact;
triggering a dynamic prediction of a new TCA trigger resulting in a TCA prediction trigger based on the TCA and the determining step;
matching, in response to determining the TCA prediction trigger, the operational TCA data to a pattern of a plurality of patterns from a TCA pattern bank, wherein the determining of the TCA prediction trigger includes determining, from the TCA data, the service level agreement (SLA) impacts exceed a target SLA impact count associated with a current threshold and determining, from the TCA data, that resource utilization exceeds a resource target utilization with the current threshold;
storing the plurality of patterns according to a tree data structure in the TCA pattern bank;
using the tree data structure to segment a total period of monitoring into a plurality of sets to establish branches and leaves as neighbors; and
calculating a predicted TCA threshold based on the pattern from the TCA pattern bank, wherein the TCA pattern bank comprises the plurality of sets.

### Claim 15 (depends on 14)

The system of claim 14, wherein the TCA data includes one or more of service level agreement (SLA) TCA data and resource utilization TCA data.

### Claim 16 (depends on 14)

The system of claim 14, wherein the TCA prediction trigger causes an adaptive thresholding process to be engaged.

### Claim 17 (depends on 14)

The system of claim 14, wherein the matching of the operational TCA data to the pattern from the TCA pattern bank includes matching based on a heuristic.

### Claim 18

A non-transitory computer-readable medium, the non-transitory computer-readable medium storing instructions that when executed a processor perform aspects comprising:
receiving threshold crossing alert (TCA) data relating to a service level agreement (SLA) and including operational TCA data;
determining that a threshold not being satisfied, the threshold associated with the TCA, is associated with a root cause of an SLA impact;
triggering a dynamic prediction of a new TCA trigger resulting in a TCA prediction trigger based on the TCA and the determining step;
matching, in response to determining the TCA prediction trigger, the operational TCA data to a pattern of a plurality of patterns from a TCA pattern bank, wherein the determining of the TCA prediction trigger includes determining, from the TCA data, the service level agreement (SLA) impacts exceed a target SLA impact count associated with a current threshold and determining, from the TCA data, that resource utilization exceeds a resource target utilization with the current threshold;
storing the plurality of patterns according to a tree data structure in the TCA pattern bank;
using the tree data structure to segment a total period of monitoring into a plurality of sets to establish branches and leaves as neighbors; and
calculating a predicted TCA threshold based on the pattern from the TCA pattern bank, wherein the TCA pattern bank comprises the plurality of sets.

### Claim 19 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein receiving TCA data includes retrieving the TCA data from a TCA database, wherein the TCA data is historical TCA data.

### Claim 20 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein the operational TCA data is monitored in a continuous manner.