Edge AI LoRa Mesh Technologies
Vermesan, Ovidiu; vorm Walde, Kai; Bahr, Roy; Conrady, Cordula; Judvaitis, Janis; Gaigals, Gatis; Karlsen, Tore; Coppola, Marcello; Sand, Hans-Erik
Chapter
Published version
Permanent lenke
https://hdl.handle.net/11250/3136231Utgivelsesdato
2023Metadata
Vis full innførselSamlinger
- Publikasjoner fra CRIStin - SINTEF AS [5801]
- SINTEF Digital [2501]
Originalversjon
Advancing Edge Artificial Intelligence: System Contexts. 2023, 1-42. http://dx.doi.org/10.13052/rp-9788770041010Sammendrag
Intelligent connectivity at the edge combines wireless communication, edge artificial intelligence (AI), edge computing and internet of things (IoT) technologies to perform machine learning (ML) and deep learning (DL) on connected edge devices. Low latency, ultra-low-energy intelligent IoT devices with on-board computing, and a distributed architecture and analytics are essential to drive intelligent connectivity.
Intelligent wireless mesh technologies exploit multiple interconnected devices, or nodes, to create a distributed network integrated with edge AI analytics using ML and DL algorithms. In an intelligent wireless mesh network (WMN), each node has embedded intelligence and can communicate directly with its neighbouring nodes and transfer data efficiently to other nodes. Compared with traditional point-to-point wireless networks, the intelligent wireless mesh approach offers several advantages, including increased coverage, redundancy, scalability and resilience.
The convergence of multiple technologies (connectivity, edge AI, IoT, distributed architectures and federated learning) delivers intelligent edge mesh communication systems that perform efficient connectivity by optimising data rates, coverage, energy, and interference.
This article overviews the latest advancements in edge AI long-range mesh technologies and applications, highlights state-of-the-art mesh communication requirements and implementations and identifies future research challenges and directions.