Granted Patent B2

US 12,574,792 · App. 17/480,529 · Granted Mar 10, 2026

Inventors: Alessandro Erta (Licciana Nardi, IT); Luca Bisti (Grosseto, IT); Carmine Benedetto (Pisa, IT)

Assignee: Cisco Technology, Inc.

H04W28/0268H04B17/318H04W4/44

Abstract

Claims (86)

1 . A method, comprising:

populating, by a mobile system, a rate lookup table using sampled data indicative of (i) signal strength measurements, (ii) quality of service tags identifying a traffic class for traffic, and (iii) transmission rates for wireless communications between the mobile system and one or more access points of a wireless network;

determining, by the mobile system, a quality of service tag for traffic to be sent between the mobile system and a particular access point of the wireless network and a signal strength between the mobile system and the particular access point;

predicting, by the mobile system and using a trained machine-learning model configured with input features comprising at least the quality service tag for the traffic and the signal strength measurements between the mobile system and access points of the wireless network, a handoff from the particular access point to a predicted target access point of the wireless network;

selecting, by the mobile system and from the rate lookup table, a transmission rate for the traffic to be sent, based on the quality of service tag and the signal strength determined by the mobile system;

sending, by the mobile system, the traffic from the mobile system to the particular access point using the transmission rate; and

in response to the handoff being predicted, selecting, by the mobile system and from the rate lookup table, a different transmission rate for the traffic during the handoff, the different transmission rate being selected by indexing the rate lookup table with both the quality of service tag a predicted signal strength predicted target access point, from entries of the rate lookup table associated with the predicted target access point.

2 . The method as in claim 1 , wherein determining the quality of service tag for the traffic to be sent between the mobile system and the particular access point comprises:

classifying the traffic to be sent by the mobile system, to associate the quality of service tag with that traffic.

3 . The method as in claim 1 , wherein populating the rate lookup table comprises:

assigning the signal strength measurements to bins, each bin representing a range of signal strength measurements.

4 . The method as in claim 1 , wherein the signal strength between the mobile system and the particular access point is a predicted signal strength.

5 . The method as in claim 1 , wherein the mobile system uses lower transmission rates for high priority quality of service tags than for low priority quality of service tags.

6 . The method as in claim 1 , wherein the rate lookup table includes different transmission rates for different quality of service tags.

7 . The method as in claim 1 , wherein the mobile system comprises a moving vehicle.

8 . An apparatus, comprising:

one or more network interfaces;

a processor coupled to the one or more network interfaces and configured to execute one or more processes; and

a memory configured to store a process that is executable by the processor, the process when executed configured to:

populate a rate lookup table using sampled data indicative of (i) signal strength measurements, (ii) quality of service tags identifying a traffic class for traffic, and (iii) transmission rates for wireless communications between the apparatus and one or more access points of a wireless network;

determine a quality of service tag for traffic to be sent between the apparatus and a particular access point of the wireless network and a signal strength between the apparatus and the particular access point;

predict, using a trained machine learning model configured with input features comprising at least the quality of service tag for the traffic and the signal strength measurements between the apparatus and access points of the network, a handoff from the particular access point to a predicted target access point of the wireless network;

select, from the rate lookup table, a transmission rate for the traffic to be sent, based on the quality of service tag and the signal strength determined by the apparatus;

send the traffic from the apparatus to the particular access point using the transmission rate; and

in response to the handoff being predicted, select from the rate lookup table, a different transmission rate for the traffic during the handoff, the different transmission rate being selected by indexing the rate lookup table with both the quality of service tag a predicted signal strength to the predicted target access point, from entries of the rate lookup table associated with the predicted target access point.

9 . The apparatus as in claim 8 , wherein the apparatus determines the quality of service tag for the traffic to be sent between the apparatus and the particular access point by:

classifying the traffic to be sent by the apparatus, to associate the quality of service tag with that traffic.

10 . The apparatus as in claim 8 , wherein the apparatus populates the rate lookup table by:

assigning the signal strength measurements to bins, each bin representing a range of signal strength measurements.

11 . The apparatus as in claim 8 , wherein the signal strength between the apparatus and the particular access point is a predicted signal strength.

12 . The apparatus as in claim 8 , wherein the apparatus uses lower transmission rates for high priority quality of service tags than for low priority quality of service tags.

13 . The apparatus as in claim 8 , wherein the rate lookup table includes different transmission rates for different quality of service tags.

14 . A tangible, non-transitory, computer-readable medium storing program instructions that cause a mobile system to execute a process comprising:

populating, by the mobile system, a rate lookup table using sampled data indicative of (i) signal strength measurements, (ii) quality of service tags identifying a traffic class for traffic, and (iii) transmission rates for wireless communications between the mobile system and one or more access points of a wireless network;

determining, by the mobile system, a quality of service tag for traffic to be sent between the mobile system and a particular access point of the wireless network and a signal strength between the mobile system and the particular access point;

predicting, by the mobile system and using a trained machine-learning configured with input features comprising at least the quality of service tag for the traffic and the signal strength measurements between the mobile system and access points of the wireless network, a handoff from the particular access point to a predicted target access point of the wireless network;

selecting, by the mobile system and from the rate lookup table, a transmission rate for the traffic to be sent, based on the quality of service tag and the signal strength determined by the mobile system;

sending, by the mobile system, the traffic from the mobile system to the particular access point using the transmission rate; and

in response to the handoff being predicted, selecting, by the mobile system and from the rate lookup table, a different transmission rate for the traffic during the handoff, the different transmission rate being selected by indexing the rate lookup table with both the quality of service tag a predict signal strength to the predicted target access point, from entries of the rate lookup table associated with the predicted target access point.

15 . A method, comprising:

populating, by a mobile system, a rate lookup table using sampled data indicative of (i) signal strength measurements, (ii) quality of service tags identifying a traffic class for traffic, and (iii) transmission rates for wireless communications between the mobile system and one or more access points of a wireless network;

determining, by the mobile system, a quality of service tag for traffic to be sent between the mobile system and a particular access point of the wireless network and a signal strength between the mobile system and the particular access point;

predicting, by the mobile system and using a trained machine-learning model configured with input features comprising at least the quality of service tag for the traffic and the signal strength measurements between the mobile system and access points of the wireless network, a handoff from the particular access point to a predicted target access point of the wireless network;

selecting, by the mobile system and using the rate lookup table, a transmission rate for the traffic to be sent based on the quality of service tag and the signal strength determined by the mobile system;

selecting, by the mobile system and using the trained machine-learning model, times to obtain the sampled data used to populate the rate lookup table in advance of the handoff being predicted, including obtaining sampled data with respect to the predicted target access point; and

sending, by the mobile system, the traffic from the mobile system to the particular access point using the transmission rate.

16 . The method as in claim 15 , wherein determining the quality of service tag for the traffic to be sent between the mobile system and the particular access point comprises:

classifying the traffic to be sent by the mobile system, to associate the quality of service tag with that traffic.

17 . The method as in claim 15 , wherein populating the rate lookup table comprises:

assigning the signal strength measurements to bins, each bin representing a range of signal strength measurements.

18 . The method as in claim 15 , wherein the signal strength between the mobile system and the particular access point is a predicted signal strength.

19 . The method as in claim 15 , wherein the mobile system uses lower transmission rates for high priority quality of service tags than for low priority quality of service tags.

20 . The method as in claim 15 , wherein the rate lookup table includes different transmission rates for different quality of service tags.

21 . The method as in claim 15 , wherein the mobile system comprises a moving vehicle.

22 . The method as in claim 15 , further comprising:

using the trained machine-learning model to select the transmission rate for the traffic to be sent from the rate lookup table based in part on a prediction by the trained machine-learning model.

23 . An apparatus, comprising:

one or more network interfaces;

a processor coupled to the one or more network interfaces and configured to execute one or more processes; and

a memory configured to store a process that is executable by the processor, the process when executed configured to:

populate a rate lookup table using sampled data indicative of (i) signal strength measurements, (ii) quality of service tags identifying a traffic class for traffic, and (iii) transmission rates for wireless communications between the apparatus and one or more access points of a wireless network;

determine a quality of service tag for traffic to be sent between the apparatus and a particular access point of the wireless network and a signal strength between the apparatus and the particular access point;

predict, using a trained machine-learning model configured with input features comprising at least the quality of service tag for the traffic and the signal strength measurements between the apparatus and access points of the wireless network, a handoff from the particular access point to a predicted target access point of the wireless network;

select, using the rate lookup table, a transmission rate for the traffic to be sent based on the quality of service tag and the signal strength determined by the apparatus;

select, using the trained machine-learning model, times to obtain the sampled data used to populate the rate lookup table in advance of the handoff being predicted, including obtaining sampled data with respect to the predicted target access point; and

send the traffic from the apparatus to the particular access point using the transmission rate.

24 . The apparatus as in claim 23 , wherein the apparatus determines the quality of service tag for the traffic to be sent between the apparatus and the particular access point by:

classifying the traffic to be sent by the apparatus, to associate the quality of service tag with that traffic.

25 . The apparatus as in claim 23 , wherein the apparatus populates the rate lookup table by:

assigning the signal strength measurements to bins, each bin representing a range of signal strength measurements.

26 . The apparatus as in claim 23 , wherein the signal strength between the apparatus and the particular access point is a predicted signal strength.

27 . The apparatus as in claim 23 , wherein the apparatus uses lower transmission rates for high priority quality of service tags than for low priority quality of service tags.

28 . The apparatus as in claim 23 , wherein the rate lookup table includes different transmission rates for different quality of service tags.

29 . The apparatus as in claim 23 , further comprising:

using the trained machine-learning model to select the transmission rate for the traffic to be sent from the rate lookup table based in part on a prediction by the trained machine-learning model.

30 . A tangible, non-transitory, computer-readable medium storing program instructions that cause a mobile system to execute a process comprising:

populating, by the mobile system, a rate lookup table using sampled data indicative of (i) signal strength measurements, (ii) quality of service tags identifying a traffic class for traffic, and (iii) transmission rates for wireless communications between the mobile system and one or more access points of a wireless network;

determining, by the mobile system, a quality of service tag for traffic to be sent between the mobile system and a particular access point of the wireless network and a signal strength between the mobile system and the particular access point;

predicting, by the mobile system and using a trained machine-learning model configured with input features comprising at least the quality of service tag for the traffic and the signal strength measurements between the mobile system and access points of the wireless network, a handoff from the particular access point to a predicted target access point of the wireless network;

selecting, by the mobile system and using the rate lookup table, a transmission rate for the traffic to be sent based on the quality of service tag and the signal strength determined by the mobile system;

selecting, by the mobile system and using the trained machine-learning model, times to obtain the sampled data used to populate the rate lookup table in advance of the handoff being predicted, including obtaining sampled data with respect to the predicted target access point; and

sending, by the mobile system, the traffic from the mobile system to the particular access point using the transmission rate.

31 . The method as in claim 1 , further comprising:

using the trained machine-learning model to select when to obtain the sampled data used to populate the rate lookup table.

32 . The apparatus as in claim 8 , wherein the process when executed is further configured to:

use the trained machine-learning model to select when to obtain the sampled data used to populate the rate lookup table.

Assignments (1)

Continuity (1)

References Cited (26)