# US 2025/0330397 A1 · Methods, systems, articles of manufacture and apparatus to manage network slices

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

- Number: US 2025/0330397 A1
- Document ID: US20250330397A1
- Status: Published application
- Kind: A1
- Application number: 19/253,160
- Filed: 2025-06-27
- Published: 2025-10-23
- Earliest claimed: 2025-04-10
- Inventors: Shu-ping Yeh; Abhinanda Dutta; Menglei Zhang; Shiva Radhakrishnan Iyer
- Web page: https://patentgrep.com/patent/US20250330397A1

## CPC

- H04L 41/5009: Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
- H04L 41/147: for predicting network behaviour
- H04L 41/16: using machine learning or artificial intelligence
- H04L 43/0888: Throughput
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Systems, apparatus, articles of manufacture, and methods are disclosed to manage network slices. A disclosed example includes accessing radio access network (RAN) data from a hierarchical network during a first time, determining a load for a plurality of slices associated with a lower level of the hierarchical network, detecting a service level agreement (SLA) violation rate associated with aggregated first predicted throughput (TP) values associated with an upper level of the hierarchical network, detecting an overprovisioning violation rate associated with the first TP values associated with the upper level of the hierarchical network, and training a machine learning (ML) prediction model with a loss function based on the load, the SLA violation rate, and the overprovisioning violation rate.

## Claims

### Claim 1

An apparatus comprising:
interface circuitry to access radio access network (RAN) data from a hierarchical network during a first time;
machine-readable instructions; and
at least one processor circuit to be programmed by the machine-readable instructions to:
determine a load for a plurality of slices associated with a lower level of the hierarchical network;
detect a service level agreement (SLA) violation rate associated with aggregated first predicted throughput (TP) values associated with an upper level of the hierarchical network;
detect an overprovisioning violation rate associated with the first TP values associated with the upper level of the hierarchical network; and
train a machine learning (ML) prediction model with a loss function based on the load, the SLA violation rate, and the overprovisioning violation rate.

### Claim 2 (depends on 1)

The apparatus as defined in claim 1, wherein one or more of the at least one processor circuit is to determine TP targets for the plurality of slices based on a difference between (a) the first TP values and (b) a global SLA TP value for the hierarchical network.

### Claim 3 (depends on 2)

The apparatus as defined in claim 2, wherein one or more of the at least one processor circuit is to determine capable TP values for the plurality of slices based on (a) the TP targets and (b) ground truth TP values for the plurality of slices.

### Claim 4 (depends on 3)

The apparatus as defined in claim 3, wherein one or more of the at least one processor circuit is to detect the SLA violation rate based on a difference between the capable TP values and a global load-adjusted SLA TP value.

### Claim 5 (depends on 4)

The apparatus as defined in claim 4, wherein one or more of the at least one processor circuit is to determine the global load-adjusted SLA TP value based on a minimum one of (a) the ground truth TP values and (b) the global SLA TP value.

### Claim 6 (depends on 1)

The apparatus as defined in claim 1, wherein one or more of the at least one processor circuit is to reduce at least one of the SLA violation rate or the overprovisioning violation rate based on execution of the trained ML prediction model during a second time.

### Claim 7 (depends on 6)

The apparatus as defined in claim 6, wherein one or more of the at least one processor circuit is to generate second predicted TP values associated with the upper level of the hierarchical network based on execution of the trained ML prediction model during the second time.

### Claim 8 (depends on 7)

The apparatus as defined in claim 7, wherein one or more of the at least one processor circuit is to determine the second predicted TP values based on load conditions of the hierarchical network during the second time.

### Claim 9

At least one non-transitory machine-readable medium comprising machine-readable instructions to cause at least one processor circuit to at least:
determine a load associated with a plurality of slices associated with a lower level of the hierarchical network;
detect a service level agreement (SLA) violation rate associated with aggregated first predicted throughput (TP) values associated with an upper level of the hierarchical network;
detect an overprovisioning violation rate associated with the first TP values associated with the upper level of the hierarchical network; and
train a machine learning (ML) prediction model with a loss function based on the load, the SLA violation rate, and the overprovisioning violation rate.

### Claim 10 (depends on 9)

The at least one non-transitory machine-readable medium as defined in claim 9, wherein the machine-readable instructions are to cause one or more of the at least one processor circuit to determine TP target setpoints for the plurality of slices based on a difference between (a) the first TP values and (b) a global SLA TP value for the hierarchical network.

### Claim 11 (depends on 10)

The at least one non-transitory machine-readable medium as defined in claim 10, wherein the machine-readable instructions are to cause one or more of the at least one processor circuit to determine capable TP values for the plurality of slices based on (a) the TP target setpoints and (b) ground truth TP values for the plurality of slices.

### Claim 12 (depends on 11)

The at least one non-transitory machine-readable medium as defined in claim 11, wherein the machine-readable instructions are to cause one or more of the at least one processor circuit to detect the SLA violation rate based on a difference between the capable TP values and a global load-adjusted SLA TP value.

### Claim 13 (depends on 12)

The at least one non-transitory machine-readable medium as defined in claim 12, wherein the machine-readable instructions are to cause one or more of the at least one processor circuit to determine the global load-adjusted SLA TP value based on a minimum one of (a) the ground truth TP values and (b) the global SLA TP value.

### Claim 14 (depends on 9)

The at least one non-transitory machine-readable medium as defined in claim 9, wherein the machine-readable instructions are to cause one or more of the at least one processor circuit to reduce at least one of the SLA violation rate or the overprovisioning violation rate based on execution of the trained ML prediction model during a second time.

### Claim 15 (depends on 14)

The at least one non-transitory machine-readable medium as defined in claim 14, wherein the machine-readable instructions are to cause one or more of the at least one processor circuit to generate second predicted TP values associated with the upper level of the hierarchical network based on execution of the trained ML prediction model during the second time.

### Claim 16 (depends on 15)

The at least one non-transitory machine-readable medium as defined in claim 15, wherein the machine-readable instructions are to cause one or more of the at least one processor circuit to determine the second predicted TP values based on load conditions of the hierarchical network during the second time.

### Claim 17

An apparatus comprising:
means for data acquisition to access radio access network (RAN) data from a hierarchical network during a first time;
means for prediction to determine a load for a plurality of slices associated with a lower level of the hierarchical network;
means for managing slices to:
detect a service level agreement (SLA) violation rate associated with aggregated first predicted throughput (TP) values associated with an upper level of the hierarchical network; and
detect an overprovisioning violation rate associated with the first TP values associated with the upper level of the hierarchical network; and
means for loss determination to train a machine learning (ML) prediction model by generating a loss function based on the load, the SLA violation rate, and the overprovisioning violation rate.

### Claim 18 (depends on 17)

The apparatus as defined in claim 17, including means for planning to determine TP targets for the plurality of slices based on a difference between (a) the first TP values and (b) a global SLA TP value for the hierarchical network.

### Claim 19 (depends on 18)

The apparatus as defined in claim 18, wherein the means for managing is to determine capable TP values for the plurality of slices based on (a) the TP targets and (b) ground truth values for the plurality of slices.

### Claim 20 (depends on 19)

The apparatus as defined in claim 19, wherein the means for managing is to detect the SLA violation rate based on a difference between the capable TP values and a global load-adjusted SLA TP value.
