Decoupling application intelligence and its orchestration on IoT devices

Thumbnail Image

Files

master_Paredes_del_Rio_Jose_2020.pdf (2.55 MB)

URL

Journal Title

Journal ISSN

Volume Title

Perustieteiden korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2020-06-16

Department

Major/Subject

Data Science

Mcode

SCI3095

Degree programme

Master's Programme in ICT Innovation

Language

en

Pages

73

Series

Abstract

The current methodology for developing applications using machine learning in IoT devices presents serious issues of scalability due to the tight coupling between intelligence (machine learning) Services on the one hand, and applications on the other hand. The main issue is the lack of life cycle management features increasing the required efforts for deploying new intelligence in devices. This thesis aims to solve this problem by developing an Intelligence Layer that enables decoupling the Intelligence Services from applications. This thesis describes the Intelligence Layer idea from the point of view of how intelligence orchestration should be performed and how to implement the intelligent functionality, taking into account communication protocols, security, availability and authorization. This research focuses on the IoT area and it introduces a local intelligence orchestration prototype. This prototype relies on technologies such as Open Neural Network Exchange (ONNX) to implement the Intelligent components, Inter-Process Communication (IPC) and Android as the desired platform. The main conclusion of the research is that the Intelligence decoupling can be achieved by the proposed Intelligence Layer. Nevertheless, the IoT scope and its applications are wide, therefore the Intelligence Layer approach should be flexible enough to adapt in order to meet various application requirements. In addition, this research is not complete due to the lack of time and available IntelligentServices to test and improve the orchestration explained on this thesis, therefore the author recommends to develop more Intelligent Services to test more cases and improve the approach.

Description

Supervisor

Främling, Kary

Thesis advisor

Ramos, Edgar

Keywords

intelligence orchestration, machine learning, IoT, data science

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202006213931

Collections

[dipl] Perustieteiden korkeakoulu / SCI