Scheduling shared data acquisition for real-time decision making

This work investigates scheduling policies for the acquisition of possibly overlapping sets of data items required to make multiple decisions by different deadlines. The work is motivated by military IoT applications in which a large number of sensors must collect intelligence data needed to make multiple decisions. For example, data from several cameras in a contested city might be needed to decide where targets of interest are. This work is based on the assumption that network bandwidth is limited, creating a significant resource bottleneck (perhaps between the sensors and the command center where decisions are made). This might be the case, for example, due to active interference by a determined adversary. A relieved sub-problem is first discussed with a corresponding optimal algorithm. Then, an improved heuristic algorithm based on the insights from the optimal algorithm of the sub-problem is presented. Finally, the new algorithm is evaluated with multiple scheduling parameters and is compared with previous heuristics, demonstrating an improved performance of our solution.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2019

Author dc:creator

• Cheng, Tai-Sheng

Contributors dc:contributor

• Abdelzaher, Tarek

• Real-time Scheduling
• Data Freshness

Statement dc:rights

• Copyright 2019 Tai-Sheng Cheng

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/105267

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/105267