École de technologie supérieure
Publish/subscribe method for data processing in massive IoT leveraging blockchain for secured storage
Abstract
dc:descriptionIn the era of the Internet of Things (IoT), the rise of Machine-Type Devices (MTDs) has brought about Massive IoT (MIoT) or massive Mache-Type Communication (mMTC), creating new possibilities for connected devices. However, dealing with the substantial and diverse data streams in real time poses challenges. To address this, we employ Publish/Subscribe messaging architecture, specifically designed for real-time data processing capabilities. Our method involves a publish/subscribe system, where device data is sent to a dedicated broker acting as a separate server. This broker allows any application to subscribe to the dataset, fostering a dynamic and responsive data environment. At the heart of our data transmission setup is a Distributed Store and Stream-Processing Platform (Apache Kafka version 3.6.1), recognized for its effective data flow management. Kafka, which takes advantage of Publish/Subscribe architecture, facilitates the connection between MIoT sensors and brokers, supporting parallel clusters for improved scalability. On top of our real-time event streaming engine, we employ analytical database software such as Apache Druid version 28.0.0, known for its real-time analysis capabilities. On the other hand, to ensure uninterrupted connectivity, we integrate a fail-safe mechanism using two Software-Defined Radios (SDR) called Nutaq PicoLTE Release 1.5 in our model. This redundancy enhances data transmission availability, protecting against connectivity disruptions. For data repository security, we leverage blockchain technology, specifically InetrPlanetary File System and Hyperledger Fabric, known for their high-performance characteristics, ensuring data integrity, immutability, and security. Our latency results show that our platform significantly reduces latency for 100,000 devices classified as MIoT to under 25 milliseconds. Additionally, our research on blockchain performance highlights our model as a secure platform, achieving over 800 Transactions Per Second in a dataset of 14,000 transactions, demonstrating its high efficiency.
Degree
thesis:*- Name thesis:degree_name
- Mémoire de maîtrise électronique
- Level thesis:degree_level
- masters
- Grantor dc:publisher
- École de technologie supérieure
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ataei Kachouei, Mohammad Hossein
Subjects
dc:subject × 1Rights
- Language dc:language
- en
Identifiers
dc:identifier.*- Identifier
- INSERT-YOUR-LIBRARY-ID-HERE3545