Systems and methods for crowdsourcing real-time mobile crowd sensing applications

Patent metadata

Field	Value
Publication number	US-12052640-B2
Application number	US-202217814445-A
Country	US
Kind code	B2
Filing date	Jul 22, 2022
Priority date	Jul 11, 2019
Publication date	Jul 30, 2024
Grant date	Jul 30, 2024

Abstract

Various embodiments of systems and methods for mobile crowd sensing applications are disclosed herein.

First claim

What is claimed is:

1. A method, comprising:
determining context requirements including requirements for a plurality of device attributes of each volunteer device of a plurality of volunteer devices, wherein the context requirements are representative of device attributes and/or available resources needed for execution of a mobile crowd sensing task, wherein the determining context attributes further includes:
determining a difference value between a context value and a context requirement value for a particular device attribute for a volunteer device of the plurality of volunteer devices;
normalizing the difference value relative to other attributes to produce a context matching index for each device attribute;
obtaining a context sense index for each volunteer device of the plurality of volunteer devices, wherein the context sense index is a weighted sum of each difference value for each device attribute;
determining a penalty value for each volunteer device of the plurality of volunteer devices based on the respective context sense index, and
minimizing an objective function such that a set of selected devices is selected which minimize the penalty value and minimize a time varying operational overhead while maintaining a minimum standard of resource availability;
obtaining context capability information including device attributes and/or available resources from each volunteer device of the plurality of volunteer devices;
evaluating the context capability information for each volunteer device of the plurality of volunteer devices relative to the context requirements for each device attribute, wherein a deviation of context capability information from the context requirements is associated with a penalty; and
selecting an optimal set of selected devices from the plurality of volunteer devices that best meet the context requirements by minimizing the associated penalty while maintaining a minimum standard of resource availability.

2. The method of claim 1 , wherein the step of evaluating context attributes is executed using expected context values, wherein the expected context values are determined using one or more stochastic models.

3. The method of claim 2 , wherein the expected context value for each device attribute for a particular volunteer device of the plurality of volunteer devices is determined by examining historical context value trends.

4. The method of claim 1 , wherein the step of evaluating context attributes is executed using current context values, wherein the current context values are empirically determined.

5. The method of claim 1 , wherein the penalty value for a volunteer device of the plurality of volunteer devices is given the value of the context sense index for that particular volunteer of the plurality of volunteer devices if the difference values are within an acceptable range and wherein the penalty value for the particular volunteer device of the plurality of volunteer devices is given a large value if the difference values are not within an acceptable range.

6. The method of claim 1 , wherein the objective function requires that a total time varying operational overhead value is less than an execution time constraint value, wherein the time varying operational overhead includes time taken for data sensing and data transfer time.

7. The method of claim 1 , wherein the objective function requires that a difference between resource availability and resource usage for a volunteer device of the plurality of volunteer devices is greater than a minimum resource limit value.

Assignees

• Gupta Sandeep
• Banerjee Ayan
• Pore Madhurima
• Chakati Vinaya
• Univ Arizona State

Inventors

• Gupta Sandeep
• Banerjee Ayan
• Pore Madhurima
• Chakati Vinaya

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• H04L67/5683

  Storage of data provided by user terminals, i.e. reverse caching · CPC title

• G06N20/00

  Machine learning · CPC title

• G16H50/80

  for detecting, monitoring or modelling epidemics or pandemics, e.g. flu · CPC title

• H04W4/029

  Location-based management or tracking services · CPC title