Systems and methods for monitoring performance of a network slice and mitigating load on the network slice

Patent metadata

Field	Value
Publication number	US-12368657-B2
Application number	US-202217809065-A
Country	US
Kind code	B2
Filing date	Jun 27, 2022
Priority date	Jun 27, 2022
Publication date	Jul 22, 2025
Grant date	Jul 22, 2025

Abstract

A device may receive network data identifying uplink/downlink packet loss percentage, uplink/downlink jitter, uplink/downlink latency, and uplink/downlink packet throughput associated with a slice of a network, and may set an uplink value for each of the uplink packet loss percentage, the uplink jitter, the uplink latency, and the uplink packet throughput. The device may multiply the uplink values by corresponding uplink weights to calculate an uplink cost, and may set a downlink value for each of the downlink packet loss percentage, the downlink jitter, the downlink latency, and the downlink packet throughput. The device may multiply the downlink values by corresponding downlink weights to calculate a downlink cost, and may calculate a total cost based on the uplink cost and the downlink cost. The device may cause another slice, with the same attributes as the slice, to be instantiated when the total cost satisfies a threshold for a time period.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a device, network data identifying uplink packet loss percentage, uplink jitter, uplink latency, uplink packet throughput, downlink packet loss percentage, downlink jitter, downlink latency, and downlink packet throughput associated with a slice of a network;
setting, by the device, an uplink value for each of the uplink packet loss percentage, the uplink jitter, the uplink latency, and the uplink packet throughput to generate a plurality of uplink values;
multiplying, by the device, the plurality of uplink values by corresponding uplink weights to calculate an uplink cost;
setting, by the device, a downlink value for each of the downlink packet loss percentage, the downlink jitter, the downlink latency, and the downlink packet throughput to generate a plurality of downlink values;
multiplying, by the device, the plurality of downlink values by corresponding downlink weights to calculate a downlink cost;
calculating, by the device, a total cost based on the uplink cost and the downlink cost;
determining, by the device, whether the total cost satisfies a threshold for a time period; and
causing, by the device, another slice or a slice instance, with the same uplink packet loss percentage, uplink jitter, uplink latency, uplink packet throughput, downlink packet loss percentage, downlink jitter, downlink latency, and downlink packet throughput as the slice, to be instantiated based on the total cost satisfying the threshold for the time period.

2. The method of claim 1 , wherein setting the uplink value for each of the uplink packet loss percentage, the uplink jitter, the uplink latency, and the uplink packet throughput to generate the plurality of uplink values comprises:
setting a first uplink value for the uplink packet loss percentage, based on a first uplink threshold;
setting a second uplink value for the uplink jitter, based on a second uplink threshold;
setting a third uplink value for the uplink latency, based on a third uplink threshold; and
setting a fourth uplink value for the uplink packet throughput, based on a fourth uplink threshold.

3. The method of claim 2 , wherein multiplying the plurality of uplink values by the corresponding uplink weights to calculate the uplink cost comprises:
multiplying the first uplink value by a first uplink weight to generate a first uplink cost;
multiplying the second uplink value by a second uplink weight to generate a second uplink cost;
multiplying the third uplink value by a third uplink weight to generate a third uplink cost;
multiplying the fourth uplink value by a fourth uplink weight to generate a fourth uplink cost; and
calculating the uplink cost based on the first uplink cost, the second uplink cost, the third uplink cost, and the fourth uplink cost.

4. The method of claim 1 , wherein setting the downlink value for each of the downlink packet loss percentage, the downlink jitter, the downlink latency, and the downlink packet throughput to generate the plurality of downlink values comprises:
setting a first downlink value for the downlink packet loss percentage, based on a first downlink threshold;
setting a second downlink value for the downlink jitter, based on a second downlink threshold;
setting a third downlink value for the downlink latency, based on a third downlink threshold; and
setting a fourth downlink value for the downlink packet throughput, based on a fourth downlink threshold.

5. The method of claim 4 , wherein multiplying the plurality of downlink values by the corresponding downlink weights to calculate the downlink cost comprises:
multiplying the first downlink value by a first downlink weight to generate a first downlink cost;
multiplying the second downlink value by a second downlink weight to generate a second downlink cost;
multiplying the third downlink value by a third downlink weight to generate a third downlink cost;
multiplying the fourth downlink value by a fourth downlink weight to generate a fourth downlink cost; and
calculating the downlink cost based on the first downlink cost, the second downlink cost, the third downlink cost, and the fourth downlink cost.

6. The method of claim 1 , wherein calculating the total cost based on the uplink cost and the downlink cost comprises:
applying weights to the uplink cost and the downlink cost to generate a weighted uplink cost and a weighted downlink cost; and
calculating the total cost based on the weighted uplink cost and the weighted downlink cost.

7. The method of claim 1 , further comprising:
processing the plurality of uplink values and the corresponding uplink weights, with a machine learning model, to generate results that determine expected uplink values for a present time period or predicted downlink values for a future time period; and
modifying one or more of the plurality of uplink values or the corresponding uplink weights based on the results.

8. A device, comprising:
one or more processors configured to:
receive network data identifying uplink packet loss percentage, uplink jitter, uplink latency, uplink packet throughput, downlink packet loss percentage, downlink jitter, downlink latency, and downlink packet throughput associated with a slice of a network;
set an uplink value for each of the uplink packet loss percentage, the uplink jitter, the uplink latency, and the uplink packet throughput to generate a plurality of uplink values;
multiply the plurality of uplink values by corresponding uplink weights to calculate an uplink cost;
set a downlink value for each of the downlink packet loss percentage, the downlink jitter, the downlink latency, and the downlink packet throughput to generate a plurality of downlink values;
multiply the plurality of downlink values by corresponding downlink weights to calculate a downlink cost;
apply weights to the uplink cost and the downlink cost to generate a weighted uplink cost and a weighted downlink cost;
calculate a total cost based on the weighted uplink cost and the weighted downlink cost;
determine whether the total cost satisfies a threshold for a time period; and
cause another slice, with the same uplink packet loss percentage, uplink jitter, uplink latency, uplink packet throughput, downlink packet loss percentage, downlink jitter, downlink latency, and downlink packet throughput as the slice, to be instantiated based on the total cost satisfying the threshold for the time period.

9. The device of claim 8 , wherein the one or more processors are further configured to:
process the plurality of downlink values and the corresponding downlink weights, with a machine learning model, to generate results; and
modify one or more of the plurality of downlink values or the corresponding downlink weights based on the results.

10. The device of claim 8 , wherein each of the plurality of uplink values is zero or one,
wherein the corresponding uplink weights sum to one,
wherein each of the plurality of downlink values is zero or one, and
wherein the corresponding downlink weights sum to one.

11. The device of claim 8 , wherein the one or more processors are further configured to one or more of:
perform load balancing for the slice and the other slice;
cause the network to switch a user equipment from the slice to a new slice based on the total cost satisfying the threshold for the time period;
cause the network to reject a user equipment registration for the slice based on the total cost satisfying the threshold for the time period; or
cause the network to switch a user equipment from the slice to a slice with less attributes as the slice based on the total cost satisfying the threshold for the

Assignees

• Verizon Patent & Licensing Inc

Inventors

• Mathew James
• Feder Peretz
• Takla Mourad B
• Vargantwar Sachin

Classifications

• H04L47/125 Primary

  by balancing the load, e.g. traffic engineering · CPC title

• H04L47/28

  in relation to timing considerations · CPC title

• H04L43/0888

  Throughput · CPC title

• H04L41/147

  for predicting network behaviour · CPC title

• H04L41/145

  involving simulating, designing, planning or modelling of a network · CPC title