Granted Patent B2

US 12,267,707 · App. 17/491,049 · Granted Apr 1, 2025

Inventors: Ratnakar Rao Venkata Rayavarapu (Bangalore, IN); Vijay Kumar Mishra (Bangalore, IN)

Assignee: SAMSUNG ELECTRONICS CO., LTD.

H04W24/08H04B17/318H04L67/12H04W8/245H04W12/06H04W28/0284G06N20/00

Abstract

Claims (65)

1. A first electronic device for optimizing processing of task requests, the first electronic device comprising:

memory;

communication circuitry; and

at least one processor coupled to the memory,

wherein the memory stores instructions executable by the at least one processor, when executed, cause the first electronic device to:

receive a task request related to an application,

identify whether the task request is capable of being processed by a second electronic device,

determine a processing mode for the task request based on the identification, the processing mode including a mode for processing, by the first electronic device or the second electronic device, the task request, and

perform at least one action based on the determined processing mode,

wherein the processing mode includes at least one of a first processing mode for processing the task request by the first electronic device or a second processing mode for processing the task request by the second electronic device, and

wherein the memory stores instructions cause the first electronic device to:

compare a first processing time for processing the task request by the first electronic device and a second processing time for processing the task request by the second electronic device,

based on the first processing time being less than the second processing time, determine the first processing mode to be the processing mode for the task request, and

based on the first processing time not being less than the second processing time, determine whether a user has selected the first processing mode to be a default mode for the task request according to a user preference or a battery level of the first electronic device is greater than a pre-defined level, and

based on the user having selected the first processing mode to be the default mode or the battery level of the first electronic device being greater than the pre-defined level, determine the first processing mode to be the processing mode for the task request.

2. The first electronic device of claim 1 , wherein the processing mode further includes a third processing mode for processing the task request by the first electronic device and the second electronic device.

3. The first electronic device of claim 1 , wherein the memory stores instructions cause the first electronic device to:

determine at least one parameter on initiating the task request, wherein the at least one parameter includes signal strength of at least one radio access network (RAN) supported by the first electronic device, type of the at least one RAN, network congestion state associated with the at least one RAN or a unique identifier (UID) of at least one application associated with the task request;

determine at least one of the first processing time of the first electronic device or the second processing time of the second electronic device for the task request using the determined at least one parameter and at least one machine learning (ML) table;

derive a cost function for the task request; and

generate a processing decision parameter based on an analysis of the cost function and the determined at least one of the first processing time of the first electronic device or the second processing time of the second electronic device for the task request, wherein the processing decision parameter indicates the processing mode for the task request.

4. The first electronic device of claim 3 , wherein the memory stores instructions cause the first electronic device to build the at least one ML table by fetching training data from at least one of a processing database of the first electronic device or a processing database of the second electronic device,

wherein the training data fetched from the processing database of the first electronic device includes information about a previously determined processing time of the first electronic device with respect to a plurality of task requests, and

wherein the training data fetched from the processing database of the second electronic device includes information about a previously determined processing time of the second electronic device with respect to the plurality of task requests.

5. The first electronic device of claim 3 , wherein the memory stores instructions cause the first electronic device to build the at least one ML table by receiving processing capabilities from at least one second electronic device.

6. The first electronic device of claim 5 , wherein the memory stores instructions cause the first electronic device to perform at least one of:

receiving the processing capabilities through a mean of crowd-sourced data from the at least one second electronic device, wherein the mean of crowd-sourced data is derived by the at least one second electronic device from a plurality of ML tables uploaded to the at least one second electronic device by a plurality of first electronic devices, or receiving the processing capabilities along with the mean of crowd-sourced data from the at least one second electronic device by performing a handshake with the at least one second electronic device.

7. The first electronic device of claim 3 , wherein the memory stores instructions cause the first electronic device to:

compare the task request and the determined at least one parameter for the task request with data of the ML table, wherein the data of the ML table includes a plurality of task requests of a plurality of applications mapped with a plurality of parameters, the second processing time of the second electronic device, and the first processing time of the first electronic device;

identify at least one task request and corresponding at least one parameter of the ML table matching with the task request and the determined at least one parameter for the task request; and

determine at least one of the first processing time of the first electronic device or the second processing time of the second electronic device mapped with the identified at least one task request and the corresponding at least one parameter of the ML table as at least one of the first processing time of the first electronic device or the second processing time of the second electronic device for the task request.

8. The first electronic device of claim 3 , wherein the memory stores instructions cause the first electronic device to:

assign at least one weight to at least one of the first processing time of the first electronic device for the task request, the second processing time of the second electronic device for the task request, the battery level of the first electronic device, or the user preference; and

derive the cost function as a function of the assigned at least one weight to at least one of the first processing time of the first electronic device, the second processing time of the second electronic device, the battery level of the first electronic device, or the user preference.

9. The first electronic device of claim 3 , wherein the memory stores instructions cause the first electronic device to:

perform an analysis on the derived cost function and the second processing time of the second electronic device for the task request, wherein the performing of the analysis on the derived cost function comprises performing comparisons among at least one of the first processing time of the first electronic device, the second processing time of the second electronic device, the battery level of the first electronic device, or the user preference based on at least one associated weight; and

generate the processing decision parameter based on the performed analysis.

10. The first electronic device of claim 2 , wherein the memory stores instructions cause the first electronic device to perform at least one of:

processing the task request by the first electronic device based on the determined processing mode being the first processing mode;

offloading the task request to at least one second electronic device based on the determined processing mode being the second processing mode; or

processing the task request using a combination of the processing by the first electronic device the second electronic device based on the determined processing mode being the third processing mode.

11. The first electronic device of claim 1 , wherein the memory stores instructions cause the first electronic device to:

download at least one computational module as library from at least one second electronic device for processing the task request.

12. A method for optimizing processing of a task request, the method comprising:

receiving, by a second electronic device, a task request from a first electronic device, wherein the first electronic device is adapted to determine a processing mode for the task request based on an identification whether the task request is capable of being processed by the second electronic device, wherein the processing mode includes at least one of a first processing mode for processing, by the first electronic device, the task request and a second processing mode for processing, by the second electronic device, the task request or a third processing mode for processing, by the first electronic device and the second electronic device, the task request; and

processing, by the second electronic device, the task request,

wherein the method further comprises:

comparing, by the first electronic device, a first processing time for processing the task request by the first electronic device and a second processing time for processing the task request by the second electronic device,

based on the first processing time being less than the second processing time, determining, by the first electronic device, the first processing mode to be the processing mode for the task request,

based on the first processing time not being less than the second processing time, determining whether a user has selected the first processing mode to be a default mode for the task request according to a user preference or a battery level of the first electronic device is greater than a pre-defined level, and

based on the user having selected the first processing mode to be the default mode or the battery level of the first electronic device being greater than the pre-defined level, determining the first processing mode to be the processing mode for the task request.

13. The method of claim 12 , further comprising:

hosting common processing modules used by multiple applications for processing task requests.

14. The method of claim 13 , wherein the common processing modules include machine learning based training models.

15. The method of claim 13 , the hosting the common processing modules further comprise hosting the common processing modules as libraries for the multiple applications.

16. The method of claim 12 , further comprising:

providing processing capabilities to the first electronic device for deciding the processing mode.

17. The method of claim 16 , wherein the processing capabilities are based on training data including a mean of crowd-sourced data, and

wherein the crowd-sourced data includes information of learning models uploaded by a plurality of first electronic devices.

18. The method of claim 16 , wherein the processing capabilities include at least one of a processing time, processing capabilities of the second electronic device in terms of load, a number of operations that the second electronic device is able to execute per second, or a number of dedicated cores available to at least one connected first electronic device among a plurality of first electronic devices.

19. The method of claim 12 , further comprising:

determining, by the first electronic device, at least one parameter on initiating the task request, wherein the at least one parameter includes signal strength of at least one radio access network (RAN) supported by the first electronic device, type of the at least one RAN, network congestion state associated with the at least one RAN or a unique identifier (UID) of at least one application associated with the task request;

determining, by the first electronic device, at least one of the first processing time of the first electronic device or the second processing time of the second electronic device for the task request using the determined at least one parameter and at least one machine learning (ML) table;

deriving, by the first electronic device, a cost function for the task request; and

generating, by the first electronic device, a processing decision parameter based on an analysis of the cost function and the determined at least one of the first processing time of the first electronic device or the second processing time of the second electronic device for the task request, wherein the processing decision parameter indicates the processing mode for the task request.

Priority Claims (2)

Continuity (2)

References Cited (32)