Abstract
Mobile Cloud Computing refers to offloading computationally intensive algorithms from a mobile device to a cloud in order to save resources (time and energy) in the mobile device. But when the connection to the cloud is non-existent or limited, as in battle-space scenarios, exploiting neighbouring devices could be an alternative. In this paper we have developed a framework to offload computationally intensive algorithms to neighbours in order to minimise the algorithm completion time. We propose resource allocation algorithms to maximize the performance of these systems in real-time computer vision applications (drop less targets). Results show significant performance improvement at the cost of using some extra energy resource. Finally we define a new performance metric which also incorporates the energy consumed and is used to compare the offloading algorithms.
Original language | English |
---|---|
Title of host publication | 2016 24th European Signal Processing Conference (EUSIPCO) |
Publisher | IEEE |
Pages | 1823-1827 |
Number of pages | 5 |
ISBN (Electronic) | 978-0-9928-6265-7 |
ISBN (Print) | 978-1-5090-1891-8 |
DOIs | |
Publication status | Published - 1 Dec 2016 |
Externally published | Yes |
Event | 2016 24th European Signal Processing Conference (EUSIPCO) - Budapest, Hungary Duration: 29 Aug 2016 → 2 Sept 2016 |
Conference
Conference | 2016 24th European Signal Processing Conference (EUSIPCO) |
---|---|
Country/Territory | Hungary |
City | Budapest |
Period | 29/08/16 → 2/09/16 |
Keywords
- Cloud computing
- Signal processing algorithms
- IEEE 802.11 Standard
- Clocks
- Energy consumption
- Approximation algorithms
- Central Processing Unit